Allan Sandage - Session I

Weart:

I know you were born in 1926, but I really don't know much else about your family. Who were your parents and what did they do?

Sandage:

My father was a professor at Miami University in Oxford, Ohio. Then he went to the University of Illinois in Urbana. My mother was the daughter of the president of Graceland College in Iowa, a reorganized Church of Latter-Day Saints now Graceland University. They met there in Iowa, under those circumstances. He was born on a farm and my mother was born in the Philippines; my grandfather was sent to the Philippines by the President (Taft) in 1909, to take over the education system of when we were involved there.

Weart:

I see. So in a way it was an intellectual background.

Sandage:

Yes, that's right.

Weart:

How did your father get into university life, starting out on a farm?

Sandage:

Well, he went to an academy in Iowa. He was the son of a man by the name of Moses Sandage, who had three boys and a girl, and he was the only one of that family that went to high school and then to college. Then he finally got a Ph.D.

Weart:

In what?

Sandage:

He was in the School of Business Administration. He was finally a professor of advertising.

Weart:

Professor of advertising? I didn't know they had such things back then.

Sandage:

Oh yes, he was not involved in science at all.

Weart:

By the way, did you have brothers or sisters?

Sandage:

No brothers or sisters.

Weart:

OK. Tell me, what was your childhood like?

Sandage:

I was raised in a family that was university-oriented; this was in southern Ohio, a small town of 3000 people, where Miami University is located. I became very interested in science during a two-year stay in Philadelphia, where my father went to work for the government in 1936-37. That interest in astronomy came from a neighborhood boy, who was also interested in astronomy, and it just grew from there.

Weart:

It was specifically in astronomy, not just in science in general?

Sandage:

In astronomy in particular, but all of science was a very strong boyhood interest. On going back to Ohio, that interest was maintained. It was a typical Midwest childhood, small town, no big city influences.

Weart:

I see. Did you make or use telescopes?

Sandage:

I did start a telescope, a six-inch telescope, but I never finished.

Weart:

That's a fairly large project to start.

Sandage:

I learned something about optics and Foucalt testing of mirrors, but didn't mount the mirror after it was completed.

Weart:

I see. Did you read a lot in your childhood? Were there any particular science books that influenced you?

Sandage:

I read a great deal. I read chemistry, I read physics, I read mathematics. I read a lot of astronomy. MATHEMATICS FOR THE MILLIONS^[1] was a favorite book. There were two chemistry books that were very important. And then the GLASS GIANT OF PALOMAR^[2] came out, and that was quite influential because it gave the history of the Hale Observatories, at that time the Mt. Wilson Observatory, in a very interesting format. That book influenced me a great deal.

Weart:

Anything else, REALM OF THE NEBULAE (Hubble), or Eddington?

Sandage:

Yes. I read Eddington’s books; I didn't understand them very well. And I read the REALM OF THE NEBULAE on my own, outside of the university, I guess about 1940, or 1941.

Weart:

Was your schooling done partly at home or mostly in regular schools?

Sandage:

It was done in regular schools. I went to high school in Oxford, Ohio, and then went to Miami University before the Second World War. I was drafted then into the Navy in 1945. But I had two years at Miami, and majored in physics. The most important person, I think, that influenced my later style of doing science was a man by the name of Ray Edwards, who was professor of physics at Miami. It was a two or three man physics department at the time. He influenced a large number of young people, including me; he had something like 100 Ph.D. students, finally, out of that small liberal arts college.

Weart:

How did he influence your style? What influence did he have?

Sandage:

Well, the style of dedication and excellence, honesty and humbleness, somehow. I can’t put it any better. He was a very demanding taskmaster, and would not settle for any hand-waving of any kind. There had to be definite proofs, in the fundamental late 19th-century style, of all the theoremsr and homework had to be precisely done. He would support those students that were very dedicated to the subject, and let the others go their own way.

Weart:

This would be in the regular teaching, labs, etc?

Sandage:

Yes, that's right. And that whole attitude toward science of precision, of no fuzzy thinking, pervaded everything he did. He needn’t say anything, it was just his style on the blackboard and in class, and it came over to the students that way.

Weart:

I see. You were living at home at this time'?

Sandage:

No. My father then went into war work in Boston, and I lived on the campus, while he and my mother were in Cambridge.

Weart:

Before you went to Miami, before you left home, what sort of feeling did people have about science in your early home life?

Sandage:

My parents were very supportive of that. I had a telescope that was bought for me by my fathery and I made sunspot observations for three or four years as a boy. He and my mother were very supportive of all those activities. They did not discourage me at all to go into that. But it was an internal drive that started at a very young age. I was nine or ten years old at the time that I knew I had to be a scientist, and in particulary I thought I knew that I wanted to be an astronomer.

Weart:

What was it that attracted you to astronomy?

Sandage:

I can’t really put my finger on it. It was just a tremendous excitement the first time that I ever looked through a telescope. That was in the back yard of this boy's house in Philadelphia. From then on, it was just a thing that had to be done. There were just no questions about it.

Weart:

Another question: did you have any formal religious training as a child?

Sandage:

Very little. The family did not go to church regularly. My mother and father were both Mormons of the Reorganized Church that stopped in Iowa. But they were not practicing Mormons. I went to the Methodist Church in Oxfordr sporadically, not in any regular way.

Weart:

I see. You didn't go to Sunday School every Sundayt that kind of thing?

Sandage:

No.

Weart:

OK. In high school, before you went to Miami, were there any teachers, or perhaps other people, who had a particularly strong influence on you?

Sandage:

Science and mathematics were always the favorite subjects. They were supportive also; they saw my interest and gave, me privileges in the lab, I’d go in and do extra work, and they would tutor me on the side.

Weart:

I see. Well, was there anything else in your early home life, or before you went into the Navy, let's say, that had a particular influence on you?

Sandage:

I think the intellectual atmosphere at home was a strong influence. It was the thing to do to get an education. And since my father had struggled to get a Ph.D., and that was unheard of in the Midwest at the time, that seemed to be the course that was laid open to me also. There was never any question in my mind that I didn't have to go to four years of college and four years of graduate school.

Weart:

Oh, that too, already.

Sandage:

That too, already. And it was not looked upon as a chore, it was looked upon as really a tremendous opportunity, and a thing that, although necessary, was a very enjoyable aspect of life. I did a lot of reading in philosophy at the time also. I first learned about philosophy from a book by Jeans called PHYSICS AND PHILOSOPHY, which is quite a remarkable book by a physical scientist. At the time, philosophy and religion were linked fairly closely in my mind, but Jeans opened up the whole world of the philosophy of science. And that interest in philosophy, the theory of science, has remained. The book by Margenau, THE NATURE OF PHYSICAL REALITY, later became a very important aspect of that type of reading.

Weart:

“Later” meaning when?

Sandage:

I think that book was published about 1950. I read it soon thereafter. The whole making of models, the validity of those models, what every scientist goes through, it's laid out there was the method by which science does develop. Those attitudes of proof and disproof, of exactness, which to through those books and go through all the training, are still very important. Especially in astronomy, were proof of anything is so hard to come by. It's the easiest science to speculate in, and the feeling of what is real and what is not, amongst the subjects in astronomy—well, there’s more philosophy in astronomy than there is in physics, where you can throw things in the laboratory and see the reactions. You have to take the experiments nature gives you, in astronomy, and try to find out what's actually going on. So the room for speculation is much greater in astronomy, and I think you have to hold a much tighter rein on yourself, to uncover truth in astronomy, than you do in any other science.

Weart:

How did this attitude develop? At what point did you begin to see this?

Sandage:

I think it was always instilled in me from the very beginning by the teachers that I had. Edwards at Miami, for example, was not a hand-waver in any way. If he couldn't derive the equations with precision on the blackboard, then he would say he doesn't know what he's talking about, and he demanded the same of all of his students.

Weart:

I see. Did you take philosophy courses'

Sandage:

I took philosophy courses after the Navyy when I came back to the University of Illinois.

Weart:

Let me first ask you about the Navy—but OK, tell me first about the philosophy courses.

Sandage:

Well, I majored in physics and mathematics and minored in philosophy. In the speculative philosophers, that is, Spinoza, Nietzsche, Schopenhauer.

Weart:

I see, not the logical positive.

Sandage:

No. That all seemed to be so clear, Hume and Descartes, their system all seemed just like science.

Weart:

There wasn't anything new.

Sandage:

Well, it was new. But that really was not talking about the root causes of things. OF HUMAN UNDERSTANDING doesn't go to the root causes, if I remember properly, so Hume didn't turn me on like some of these others.

Weart:

I can appreciate that. It's in a way similar—or to the impulse to study science. Is there also an interest in the root causes there?

Sandage:

It's all part of the same thing: a world system has to be true, in the sense that the universe is only one way. And science is the way to uncover it. I understand your dilemma now as to what I've said, because Schopenhauer and Nietzche were more philosophers of the mind.

Weart:

No, it's not a dilemma at all. It seems to me it might represent a similar impulse, to look at different ways of trying to understand the whole system.

Sandage:

Yes.

Weart:

Is that correct? When you were doing it at the time, did it seem to all fit together?

Sandage:

The question of mind and matter—they were both aspects of the universe. Matter was the aspect that science could uncover precisely, unequivocally, with no questions ultimately. The road to get to that positive statement, or to positive statements about science, was very rocky and clearly full of mistakes, but there was a final solution, if you want to put it that way, like Newton’s equations or Einstein’s equations. The physical universe is the way it is. And the question is, is the mental universe that way also? The system of thought by these speculative philosophers were attempts to find that out. I don't know the answer to that second question.

Weart:

Are you still interested in that?

Sandage:

Oh yes, I'm still interested. Periodically I go back to that. But it seems more and more that those questions are on much shakier ground. Those systems are built on jello. But the mystery still existsy as to the relationships of mind and matter. I still don't real I y understand that.

Weart:

All right, none of us do. But back to the Navy, then, tell me about what you did in the Navy that may have had later meaning for you?

Sandage:

I was in the Navy 18 months, during the last years of the Second World War. I went into the electronics maintenance program. A large number of people now in science were in that program, a large number of astronomers that I saw there. Art Code was in ity for example, and Al Wilson, who later did the Palomar Sky Survey.

Weart:

You met them there?

Sandage:

I met them there, that's right.

Weart:

Where was this located?

Sandage:

In Chicago at first, and then in Gulfport Mississippi, for three months of intermediate schooling, and then nine months on Treasure Island in San Francisco. This was the electronics program, repairing radios and radar. I came out as an electronics technician's mate.

Weart:

Did you ever use that?

Sandage:

Oh, a little bit. I think my extreme dislike of electronic was generated by that experience (laughter). But also, that furthered my education in engineering. The Navy courses were excellent, so it was an adjunct to school.

Weart:

I see. Tell me how you chose the University of Illinois.

Sandage:

My father persuaded me to go there. He had transferred after he got back from government service, from Miami to University of Illinois. So I spent the last two years living at home after the war.

Weart:

Your parents had moved there.

Sandage:

That's right. There was an economic problem, and living at home and with goverment benefits from the GI Bill, it was easy to go to school; in fact it was probably the only way I could have gone to school. I majored in physics. F. Wheeler Loomis was the head of the physics department at that time. Those two years, I thinkt were amongst the most important formative years of my life. To go from the one-many Ray Edwards, physics departmenty to a full-flown department of physics at a major university, where many of the people had been involved in war work—radar and the atomic bomb, and then to be in amongst 30 or 40 physics majors—the pressures were great, and the drive for excellence tremendous.

Weart:

Were a number of these people older than you, because of the GI Bill?

Sandage:

I don't think so. I think there may have been an age spread of three or four years, but that's not a strong memory. The strong memory was of the toughness of the courses and the expectations that the professors gave to the students. The most important people for me at that time were (G.M.) Almy, who was the professor of optics at Illinois; (Gerard) Kruger, who was professor of modern physics; and above all, Robert Becker, who was a Cal Tech graduate and taught analytical mechanics. I couldn't solve the problems the first month or so, and he was very helpful and patient, and taught me the methods to solve all of the homework problems.

Weart:

You’d go into his office, that sort of thing?

Sandage:

I'd go into his office and tell him that I'd worked 15 hours on this problem set and hadn't gotten very far, and could he show me? And he would do that. Then, electricity and magnetism was taught by A. O. Hansen, the inventor of the klystron. I saw Donald Kerst there, although he was not one of my teachers.

Weart:

That was quite a physics department, wasn't it?

Sandage:

It really was incredible.

Weart:

Did you get a very modern, up-to-date education, do you think?

Sandage:

I think I got a very good education in physics.

Weart:

Did you have quantum mechanics?

Sandage:

I did not have quantative mechanics, no.

Weart:

Or relativity?

Sandage:

I did not have relativity.

Weart:

Interesting, very few schools seem to have been teaching those things to undergraduates at that time.

Sandage:

That's right.

Weart:

So what you had was E & M and optics, that sort of thing.

Weart:

And analytical mechanics, a classic course. Oh, and there was a man by the name of (James) Bartlett, who was a theoretician. I didn’t understand any of his lectures at first. He was really a theoretician in the elegant sense of Hardy in pure mathematics (now looking back on it—I haven't thought of this for years). He's famous of course for the Bartlett potentials, in the early days of quantum mechanics.

Weart:

What about lab courses, did you have much of that?

Sandage:

I had courses in electricity and magnetism and optics. The optics lab course was extraordinarily interesting. They had us do interference experiments and microphotometer the photographic plates and calculate the various patterns of diffraction.

Weart:

A photoelectric microphotometer?

Sandage:

Yes, that's right. The signals were recorded on photographic plates, so we learned how to make photographs and develop them.

Weart:

So you went into the darkroom. Then you measured them photoelectrically. That's a very useful education.

Sandage:

Yes. But then I had a parallel career as an undergraduate, because I majored in astronomy also.

Weart:

So you had a dual major in physics and astronomy, and a minor in philosophy?

Sandage:

It was a major in physics and mathematics, but I worked at the observatory, and had a minor in philosophy. I didn't take any formal astronomy courses in the elementary sense, because I'd known al I that before.

Weart:

From your reading?

Sandage:

That’s right. But I did take analytical mechanics from Baker, orbit calculations.

Weart:

Celestial mechanics?

Sandage:

That's right. I started a junior-senior research problem under (Robert H.) Baker, with the small camera at Illinois. That training later came in very handy here at the observatories when I came to Cal Tech, because I had learned under Baker how to transfer magnitude sequences from one area of the sky to another, by photographic transferst and by step-scale photometry, and the Argelander method. So I knew all of that technique before coming to Cal Tech.

Weart:

I see. How did you happen to do this? Were you employed at the observatory?

Sandage:

No, I volunteered.

Weart:

You just went to the observatory and saidy "I want to do something?''

Sandage:

That's right. And he was plugged into the star-counting circuit that Bart Bok had organized from Harvard. Illinois was part of the star-count network. So Baker had me do an area, which was the area in Perseus—get the plates, transfer the magnitudes, count the stars, do the analysis for the density and absorption characteristics in that region.

Weart:

I see. You did all this as a volunteer.

Sandage:

That's right.

Weart:

Why didn't you major in astronomy, then?

Sandage:

Well, I figured that if I wanted to be an astrophysicisty astronomy was very easy, and physics was very hardy and it was the physics I needed.

Weart:

Where did you get the notion of astrophysics?

Sandage:

It was just in the air, in my boyhood, I guess. But I really didn't understand what astrophysics was until coming to Cal Tech and taking the first few courses from Greenstein.

Weart:

But you already understood that you would need the physics to do astronomy. That's interesting. What was it like starting to observe at a real observatory?'

Sandage:

Here?

Weart:

No, at Illinois?

Sandage:

Well, it was cold. Those midwest winters are incredibly cold, and Perseus is in the winter sky.

Weart:

Tell me about the other science students at Illinois. Did you know where they came from and what sorts of jobs they were aiming for?

Sandage:

That memory is dim. There are a few people that I remember, but I must say that I can remember my social contacts and my physical surroundings hardly at all at Illinois. And that's surprising to me. I remember a few professors. I remember how difficult the courses were. I remember how hard the work was. WEART; You remember the professors pretty well, actually. Were there any fellow-students that you kept up contact with later on?

Sandage:

Yes, there were a few. There’s Walter Connally, who is now in North Carolina. He's head of the physics department there. And there was George Bradly, who was head of the physics department, at I think, Western Michigan University.

Weart:

By the time you graduated, you were quite certain that you were going to go on in astronomy.

Sandage:

Yes.

Weart:

What sort of life did you expect to lead as an astronomer? Did you see yourself doing research, or teaching, or what?

Sandage:

Oh, I knew I had to do research. This project at Illinois, star-counting, clearly outlined for me what research was like, and it was everything that was interesting, exciting, challenging, to set a problem and then go about solving it. That was the first experience in a large research project, and it certainly was an indication that that's where the future lay.

Weart:

I see, even though star-counting doesn't have particular rewards for the person who does one patch of the sky.

Sandage:

I'm doing that now.

Weart:

Yes, but now it's your project and you're doing it with a definite goal.

Sandage:

But I figured at the time that it was a learning process, and Baker left me completely by myself to understand the methods, to understand how to develop plates. I didn't know how to develop plates, really, up until that project.

Weart:

What sort of job prospects did you expect, in the 1950s.?

Sandage:

I didn't worry about that. I knew I had to be an astronomer; the future would take care of itself if I trained myself sufficiently. I clearly wanted to be some place with a big telescope. But at the time, I didn't think of galaxies or cosmology or stellar evolution. Stellar evolution had not been invented at the time, in fact.

Weart:

Or at least it wasnyt something that they discussed in astronomy courses.

Sandage:

That's right. The H-R diagram was the vehicle in stellar astronomy, and you used it in many waysy but up until 1948 or so, there were no indications really of what it was about.

Weart:

Yes, they had no idea which way the stars moved through it.

Sandage:

Yes, or the age dating possiblities, or the fact that from the H-R diagram you cap understand the history and formation of the various substrata of the galaxy.

Weart:

I understand. What was your family's attitude toward your choice of career?

Sandage:

Again, they were very supportive, and they realized that some career in science was what interested me, and they were encouraging. So I applied to Cal Tech and I applied to Harvard, for entrance to the graduate school in astronomy.

Weart:

I see, you knew enough from talking to people or whatever to know these were the two places to apply.

Sandage:

That's right. Bart Bok came through Illinois, and in fact invited the two of us that were in the astronomy department back for a summer at Harvard. This was in 1947 or 1948.

Weart:

I see. Who was the other one?

Sandage:

That was a man by the name of Tom Jones. He was a mathematician who was also involved in the star-counting project.

Weart:

So you did go back to Harvard?

Sandage:

We both went back to Harvard for the summer school and spent the summer out at Agassiz Station.

Weart:

Observing?

Sandage:

Observing, that's right.

Weart:

It was one of these great summer schools they hady where all sorts of famous people were lecturing?

Sandage:

No, this was a summer school for the observing students. We lived on the grounds, and Bok would come out from Cambridge. I guess he lived there for two or three weeks of the eight or nine. There were many other students there at the time; Arthur Hoag was there, who’s now director at Flagstaff, Bappu was there, "Chip" Arp was there.

Weart:

Coming from various places?

Sandage:

That’s right. (Isadore) Epstein was there, who's now professor of astronomy at Columbia.

Weart:

This was the first time you'd really been part of an astronomical community.

Sandage:

That’s absolutely right, yes. It was a very great intel I ectual experience.

Weart:

Was this the first time you were exposed to some ideas about astronomy? evolutiony or galaxies or whatever?

Sandage:

No. That again was essentially mapping various segments of the Milky Way, setting up magnitude sequences, finding variable stars. That whole aspect of a deeper understanding didn't come over in any lecture series or courses that were given at that time. The big picture of trying to understand our galaxy in the context of the universse was not talked of in those times. I think that there was a real change in the deepness of interpretation, that was fermenting at this time. But what is now taken for grantedy 25 years down the line, was really not in ferment at that particular time, 1947-48.

Weart:

This is a feeling I get from other interviews too, that people have told me. So these were things that you would pick up later from talking to people or from the journals?

Sandage:

No, those were things that developed here at Mount Wilson and Palomar Or not here exclusively, but things that developed throughout the country, in almost a spontaneous flowering, in a five-year interval. The evolutionary ideas were around from the time of Opik in 1930, when the stars moved off the Main Sequence. But the tying together of what happened after the Schoenberg-Chandrasekher limit of 10 per cent fuel burned was reached; this did not come until—

Weart:

–until you did it.

Sandage:

No, until Schwarzschild did it.

Weart:

We'll have to talk about that.

Sandage:

Sure. But I think that bringing together of two series of ideasp from theory and from observation occurred after 1950.

Weart:

I understand. Sop how did it come about then that you went to Cal Tech rather than Harvard?

Sandage:

Well, I guess I was interested in the 200-inch telescope. We'd also come out, my parents had come to California in 1941 in the summer. My father had a teaching position at Berkeley and we lived in Berkeley for the summer, and we visited Mt. Wilson on the way to Berkeley. We visited this office building at the time and I visited the optical shop. Then we went up to Mt. Wilson, and there was a tremendous desire to be associated with the Mt. Wilson Observatory. And the only way to do that was to come out as a graduate student at Cal Tech.

Weart:

Fortunately they'd just started to accept them.

Sandage:

That's right. I actually applied in physics, because I didn't know that there was an astronomy school. I went in to Wheeler Loomis and asked if he’d write a letter of recommendation, and he did, right on the spot with me in the office, dictate a letter to DuBridge. Then I was accepted. The letter came saying that the graduate school in astronomy was just beginning and "You have now been accepted in that school at Cal Tech.''

Weart:

I see. It was stated in Loomis' letter somehow that what you were really interested in was astronomy?'

Sandage:

That's right.

Weart:

How were you supported as a graduate student?

Sandage:

I worked 15 hours a week. I had $900 a year from that, and then I had a tuition scholarship.

Weart:

Tell me something about the teaching at Cal Tech, Greenstein and so forth, what your courses were like.

Sandage:

Greenstein was the sole professor of astronomy for two or three years. And he was a great man; he is a great man. He came out from Yerkes to begin a school of astronomy at Cal Tech, with Santa Barbara Street standing aloof, in essencey so that he had to begin on his own. And he did it, I guess with five studentsy teaching all the courses. He taught stellar atmospheres, stellar interiors, the interstellar medium, practical astronomy, astronomical methods of observation, the whole works. I don’t see how he did it.

Weart:

Did people come down from Santa Barbara Street to teach?

Sandage:

Occasionally, but the style of the courses was much different. Greenstein taught a problem course, like the Physics problem courses at Cal Tech. People up Santa Barbara St.) taught relatively qualitative courses, more descriptive courses. They were very important, but not astro-physics.

Weart:

They would lecture but they would not necessarily grade problem sets and so forth?

Sandage:

That's right. But we had problem sets from Greenstein every week, in atmospheres the first year and in interiors the second year; that was very high-level stuff. It was essentially Unsold's book on stellar atmospheresp not translated.^[3]

Weart:

In German?

Sandage:

In German, but Unsold's German style is fabulous. it just flows. Itys not complicated German, and most of the book is equations anyway.

Weart:

I saw a translation of that, made up by students at Harvard or whatever.

Sandage:

Yes, there was a translation two years later by Keith Pierce, from Michigan.

Weart:

But that came a litle too late to help you.

Sandage:

Yes, that's right. Then later Unsold came and lectured. So Greenstein also brought in very famous people. Von Laue was here, in the astronomy department; Cowling came; Hoyle came

Weart:

These would be for seminars or whatever?

Sandage:

For semesters, and they would give courses. Stromgren came; I remember a course on what was the beginning of stellar evolution from Stromgren. So under Greenstein, together with the courses in the physics departmenty we had a good grounding in theoretical astrophysics.

Weart:

I see. Did you have any courses in cosomology or extragalactic work?

Sandage:

Nor none. I'd read THE REALM OF THE NEBULAE and that was the only source outside the ASTROPHYSICAL JOURNAL literature that was available.

Weart:

That was your first introduction to ideas of the expansion of the universe and so forth?

Sandage:

That's right. Hubble did come down and give a lecture course on galaxies per se, but not on the expansion.

Weart:

When did you first encouner the idea that the expansion V±5 an expansion? I suppose that was in THE REALM OF THE NEBULAE as a possibility.

Sandage:

Well, I think more than a possibility. I never did believe, or I never was exposedy to the idea that it was not an expansion.

Weart:

When did you first hear of the steady-state theory?

Sandage:

Essentially at the time that it was expounded, and it was through the Sunday supplements. There was a piece, and it said that Fred Hoyle was to come to Cal Tech and try to convince the astronomers that the universe really was in a steady state.

Weart:

I see. How did you find out about it scientifically?

Sandage:

Hoyle came and gave a series of lectures.

Weart:

So you heard about it from the horse's mouth, so to speak.

Sandage:

I was also then an informal student of both Baade and Hubble. I had very strong interactions with Santa Barbara Street.

Weart:

I guess, before we get into that, I should ask you a little more about the teaching, but let's follow cosmology for one minute. What was the attitudep I can't say at Cal Tachy that's just Jesse Greenstein—what was the attitude at Santa Barbara St. then towards cosmology?

Sandage:

You know, cosmology at that time was a one or two man subject, in the world. And those two or three people were Hubble, Humason, and Baade. Almost all the other astronomers every place else in the world were doing something else. That's hard to understand now, because the emphasis at the moment is so strongly cosmological. If you pick up any issue of the ASTROPHYSICAL JOURNAL, half the issue every month is on cosmology. But that was not the case. It was brought about in the 1950s this strong monopoly with other three or four people in the world being involved, by the accidents of the large telescope. I'd say Mayall was also one of those few people. And then there were the theoreticians—Eddington and Friedman, H.P. Robertson, and company—but they were not in the mainstream of the observational effort was a very small y was not a major subject of emphasis.

Weart:

I've always been interested, why was so little done in cosmology between 1930 and 1950?

Sandage:

Because there was only one large telescope, and that was the 100-inch on Mt. Wilson. That was the only telescope by which any of the information could be obtained. Lick had the 36-inch Crossley and the 36-inch refractor, and they couldn't do very much concerning galaxy counts in depth. At Harvard, Shapley had done some things on the distance scale, star counts in the nearby galaxiesy galaxy counts to the 17th or 18th magnitude, there was a long program going at Harvard. But that was in a mapping sense, not in the sense of trying to understand origins and evolution.

Weart:

It seems that not only was there a lack of data but a lack of interest, in that period, or was there?

Sandage:

I guess there were so few astronomers. You'd go to the American Astronomical Society meetings, and there were maybe 300 peopley and that was the extent of the total American effort in astronomy. There were only 300 astronomers and everyone knew each other. Curtis was a very important man in the early days, clearly.

Weart:

Much earlier, of course.

Sandage:

That's right. In the early days of beginning photography, there were people like Keeler and Curtis. But then Hubble really did dominate the field, from 1925 to 1950.

Weart:

Do you think it was just that no one felt that they could compete with Hubble and the 100-inch?

Sandage:

They didn?t have any telescope.

Weart:

They didn't have any way of competing.

Sandage:

No way of competing, that's right. He controlled the entire means of production.

Weart:

OK. Getting back to the instruction here—what about the other students of astronomy here? How did you all interact?

Sandage:

We were all very supportive of each other. The courses were difficult; we were difficult; we were the beginnings; and I think that the students and Jesse Greenstein were a close-knit group. One had Santa Barbara St., that helped a bit, but it was a great colossus up here (at the Mt. Wilson Office), and the entrance was not so easy for the students to just come up. Then there was the enormous physics department; that was friendly, but Greenstein had to start a new department. And there were pressures for excellence, because the 200-inch was there and the opportunities were there. So we all understood that we were in at the beginning. Those people that were in that first class—there was Helmut Abt, who's now editor of the ASTROPHYSICAL JOURNAL, Morton Roberts, whoys a very good radio astronomer at NRAO^[4] and Jim Parker—I don't know what has happened to him now; he got his Ph.D. in solar astronomy, and as far as I know is out of astronomy. Then, the second year, Chip Arp came–gosh, I can't recall.

Weart:

Well, there are records. Did you sort of get together with these people and solve problems?

Sandage:

Oh, sure. We'd take classes together we'd try to do the homework together, and it was a close-knit group.

Weart:

Did you students ever get together to ask for a particular course to be given, or a seminar?

Sandage:

No. The curriculum was pretty well set, and most of the courses were in physics. We were required to take the core courses in physicsy and in addition to that, the three required courses in astronomy: stellar atmospheres, stellar interiors, and the interstellar medium.

Weart:

I see. Did you ever have any formal instruction in general relativity theory?

Sandage:

I took a course in relativity from H.P. Robertson.

Weart:

I suppose at this time youyd get quantum mechanics also.

Sandage:

I had quantum mechanics under Feynman. I had analytical mechanics under Leverett Davis. I did not take the electricity and magnetism under Smyth.

Weart:

OK. The next question is about getting into research. I notice, even aside from Sanata Barbara St., your first paper was on the solar excitation temperature of vanadium I^[5] ——

Sandage:

–that's right, you've done your homework.

Weart:

I always check the first paper——

Sandage:

–I see (laughter)—

Weart:

Nobody forgets their first paper.

Sandage:

Well the course by Greenstein in stellar atmospheres was fascinating. Greenstein is a remarkable teacher. He's the most disorganized teacher in the world, and if you're interested in the subjecty you know you have to go out and learn it on your owny with the lecture notes having to be reorganized and understood. So after every lecture, every student would go and rewrite the notes, organize the notes, go to the literature, understand what he saidy and get prepared for the next day's lecture. In that way, with Greenstein's help and disordery we really learned the subject. The thing that I remember about that course was the magic of determining abundances of the chemical elements through spectroscopy. Now, in ordinary laboratory spectroscopy, you just take ratios from known samplesy and that's nothing. But here, You have an object like the suny where you can't go and get your standards, and you have to understand all about the atomic f-values and equivalent widths. The curve-of-growth was a magic vehicler when it was taught. And I had to do something in that field, to understand it. There's no better way than to do a problem with it, so this excitation temperature, curve-of-growth, Boltzmann factorsy was a good learning procedure.

Weart:

So you volunteered to do it?

Sandage:

Well, I had the summer off. It was an interesting thing. Something happened so that I couldn't do what I was assigned to do (see below). So I had six weeks and I had to find something to do. So I just decided to do thisy on my own, and when he [Greenstein] came back; I showed him the paper. It was not in very good form, clearly; it was overwritten. He taught me how to write it. And that’s the origin of the paper. It was a summer research project.

Weart:

I see. I noticed you used Munch?s model. Was Guido here yet'?

Sandage:

Guido wasn't here yet.

Weart:

It's just a model that you picked up from the literature, or from Greenstein.

Sandage:

I guess that's right. I think Guido came the second or third year, but he was not here the first year, and this paper was written in the first summer.

Weart:

OK. Then tell me about getting in touch with Hubble and Baade, and starting to work on the 200-inch.

Sandage:

Well Hubble asked Greenstein if he had a student who would be willing to work; he had a project, and wondered whether Greenstein could send somebody up. So he sent me up. I don't know why he sent me up. He could have sent any of the other four. So I came up, and Hubble said, "I want, with the new equipment on Palomar, to make more precise the log N(M) relation, the numbers in depth of galaxies. What I want to do is, with the 48-inch Schmidt, which has capability of making squares, schraffier-kassettes, determine the magnitude system well enough so that we can get the distribution of galaxies in depth. Would you be interested to do that?" I said, "Sure." So he got the plates and got me started on the analysis. He transferred from selected areas, where the magnitude sequence was determined, to various areas of the sky. There were four areas of the sky, and he wanted to take the counts to the 18th magnitude by actual measurementy not estimation. Then he went off for the summer. This was the summer of 1950. Oh, that's what happened in the summer of 1950, when the first paper on the solar thing was done. I was to work for Hubble. Hubble had a heart attack then while he was awayp and there were unexplainable differences in the transfers to the selected areas, of 3/10ths of a magnitude, that I thought were large enough so I couldn't continue on the project till I got some instruction. So that's where those six weeks of interim came, when I did the solar work at Cal Tech. Then he (Hubble) came back, and in the long process of recovery, he still wanted to continue his work with the 200-inchr starting to collect plates in the M81 group for Cepheids. It then fell upon an assistant to get the plates for him, because he couldn't do it.

Weart:

I see. Before this he had taken the plates himself?

Sandage:

That's right.

Weart:

I think it would be very interesting to hear how you learned to use the 200-inch. It would give a picture of the 200—inch as it was during its early days.

Sandage:

When the 200-inch went into operation, the first scheduled run was November 1949p and Hubble was the first scheduled observer. Plates had been taken during the testing period by Bowen before that, so there had been plates for four or five months, taken by Bowen. But Hubble observed from November 1949, to about June 1950, and that was the total extent up until the time of his heart attack. Then he had this massive heart attack while he was fishing in Colorado, and was out of action for about a year. In the interim period I was sent up to take the plates for him at the telescope. Humason was the one that checked me out on that. I'd had observing experience before at Illinois, but clearly not on any large equipment and I had had observing experience with the 60-inch on Mt. Wilson, during the first summer.

Weart:

How did that come about?

Sandage:

I came up with Chip Arp and Walter Baacle; it went through Greenstein. We wanted to observe globular clusters, the population II Cepheids in globular clusters. Greenstein contacted Baade, and he said, "Sure, I'll check these two people out." So Baade went to the mountain with Arp and me and spent seven nights with us, on the 60-inch, showing us how to observe, first, and then all kinds of tricks. First, the simple Foucault test, but not only -that, how to tell from the nature of the cutoff of the mirror whether you are on the optical axis or not, whether the mirror's in good shape, whether you have a lot of coma in the center of the field if it's not in good shape—

Weart:

Just by using a knife-edge on a star.

Sandage:

That's right. How to locate the optical axis by the symmetry of the coma images, all that stuff. So I already knew how to observe with a small large telescope, and I guess it was a matter of taking whoever was available to carry on Hubble's program when he couldnyt do it.

Weart:

I'm interested to know more of the details, because this is all sort of an oral tradition, oral instruction, how to use a big telescope, which I don't suppose is the way they're used very much any more.

Sandage:

Well, there are some people who still take photographs.

Weart:

Right. But it would be interesting to know a little more about the details, because I don't suppose it's really written down anywhere.

Sandage:

It's not written down.

Weart:

Tell me a little more about it.

Sandage:

Especially on Mt. Wilson, which is an older establishment, in the sense that the way to get to the focus is archaic. You have the Newtonian platform, which has to be in place, and the telescope is moving continually, so you have to move the dome and the Newtonian platform at the same time you are attempting to get the long-exposure photograph. Then, the process of guiding. You guide on a coma imagey because you're quite off-axis with the guiding eyepiece. The images are not roundy and Baacle taught us how to bisect the coma image with the cross-wires, putting the intersection of the cross—wires on the tip of the coma image, where the object is brightest. And trying to get guide stars that are in the cardinal directions, so the bisection of the fan of the coma is east-west or north-south. And since the guiding is to compensate for rates in right ascension, then you usually have to do it east-west.

Weart:

The guiding is mainly for one-direction?

Sandage:

If the axis of the telescope is precisely on the pole, then there will be no declination guiding, except for differential refraction as you rise. If the telescope axis is misaligned with the pole, then there will have to be guiding in both directions.

Weart:

But it usually was well enough aligned?

Sandage:

That's right. But to complicate this, there is a periodic error in the drive screw of the 60-inch, which is 1 1/2 seconds of arc in amplitude and 80 seconds of time in period. Baade didn't tell us that. He wanted to see whether we would discover thatr whether we'd push the east button, then the west buttony then the east button, then the west button. He could tell by listening to the relays click and and the time period whether we were guiding well or not. So all those subtle things—in seven days we got a very thorough indoctrination. And he taught us about plate combinations, with filtersy to get the standard wavelength bands. There were sufficient cloudy nights that he then taught us about the populations, population I and II.

Weart:

I see. That's when you learned about it—but you must have already known?

Sandage:

Nor this was the first year. I'm a little hazy about exactly when—

Weart:

You said you and Arp wanted to study the population II Cepheids in clusters. You wanted to study Cepheids in clusters, you didn't know those were population II?

Sandage:

I can't remember exactly how all that came about. I guess that must have been the second year. I guess the chronology is this: I came in September 1948. I came to Santa Barbara St. to work for Hubble in the summer of 1949. Arp came as a graduate student in 1950. We then contracted Baade in 1950, and I must not have gone to the 200-inch until 1951. I can check immediately. (Gets logbook)... The first observations I made with the 200-inch were in September, 1951.

Weart:

I see, you have it in your log.

Sandage:

So Hubble must have had his heart attack in the summer of 1950, and we'd been there two years. That's how it happened. I had the training on Mt. Wilson with Baade during the second year.

Weart:

I see. What was the relationship with the night assistant like on Mt. Wilson, what did he do?

Sandage:

Well, he's the one that sets the telescope, and then makes all the records, and to the younger people, tells them what to do. If they're hitting the platform or about to hit the platform—he'll train them.

Weart:

I see. He makes sure that you don't run the telescope into something.

Sandage:

That's right. He's really an expert in holding the youngster's hand. The night assistant that we had was Gene Hancock, who came in 1946. He worked with Hubble and Baade. He took over from a man that had been there from the beginning, Tom Nelson was his name, who was a great tradition at the observatory. Tom Nelson is still alive at the age of 98. He retired at 65, having been there from the beginning, 1910 or so. So Hancock was at the training sessions that Baade had with Arp and myself at the 60-inch. He has been on Mt. Wilson for 25 yearsy helping astronomers all the time. After setting the telescopey he would keep the time recorclsy tell you when the exposure was over—-

Weart:

I see. So all you had to do was guide.

Sandage:

You had to focus the telescope also. The plate glass mirrors on Mt. Wilson change very greatly with temperature. The gradient of focus change is very high, and you have to focus after every exposure, or before the new exposures. You do that by finding a bright star and doing a Foucault test. You can tell the nature of the cutoff, and the seeing; you can tell whether you're on the optical axisy and you can tell the focus.

Weart:

This must make trouble if you're running a very long exposure.

Sandage:

Well, that's the problem with the 60 and 100-inchy and that's what Baade had to overcome when he resolved the Andromeda Nebula. You know what the whole statement of his in his resolution papery on how the conditions had to be just righty and the change of focus had to be done during the exposurny and you can do that by looking at the nature of the coma image. The seeing has to be very good for thaty but you can see striations on the coma image. All of that technique was started with us, by Baacle; it took many years to really understand all of the idosyncracies of the telescopes and the problems of the seeing, but Baade was th one who started the training.

Weart:

I see. Now tell me about learning to use the 200-inch.

Sandage:

Humason took me down the first time, and was the one who checked me out in the prime focus cage. The seeing was very good that first night. Hubble had control of the program, in the sense that before every run he would decide which objects were to be taken, and there were three lists of objectst one for very good seeing, one for intermediate seeingy and one for poor seeing. So you had to use your judgement as to which program to go on, according to the conditions. That’s still true for every astronomer: you go to the mountain with three or four programs, always depending on the conditions. So we did a Foucault test, and did a series of focus plates, because there are systematic personal differences of the eye, between where the knife edge is set, which is par focus to the plate holder, and where each observer says there's a cutoff. We took the first hour or so and did a series of focal studies.

Weart:

Finding your personal equation, so to speak.

Sandage:

That's right. Humason stayed up for two nights. The 200-inch arrangements are much easier than the 60 and 100-inch on Mt. Wilsony so everything worked quite well.

Weart:

What was it like?

Sandage:

Oh, it was fabulous. I can't really reconstruct the first three or four months of that period. So much was happening. First of all, it was an opportunity that was beyond any imagination, observing with the 200-inch, and secondly, working on the long range programs of cosmology with Hubble. And at the same time being a graduate student, trying to pass the courses in physics and in astronomy—so it was a very high-pressure atmosphere. The work on the mountain was an escape, but you knew that your sins would catch up with youy because you were four days away from campus and courses, and you were pretty well swamped. I’d bring the plates back and have a consultation with Hubble. He would then decide what to do next. At this time he was trying to find Cepheids beyond the local group.

Weart:

Were you looking for bright variables?

Sandage:

Maybe you’re thinking about the paper that was written with Hubble^[6] on the bright variables of M33?

Weart:

Yes.

Sandage:

That was done before the observations were started at Palomer before his heart attack. That was a kind of a test. He had been taking plates on Mt. Wilson of M33 and M31 since 1920, and had accumulated plates up to 1948, and on those platesy which he blinked every time he came down, very bright blue irregular variables had been found, but he had never worked them up. So my chronology is all wrong: before I started to work for him that summer, I came up and worked on these M33 plates.

Weart:

Blinking the plates?

Sandage:

No. He had already located the bright blue variables, but there were no magnitude sequences. The training with Baade had already occurred, with the 60-inch, so I knew how to observe with the 60-inch, and I told him (Hubble) I could get magnitude sequences, by transfer, with the 60 inch.

Weart:

Right, and you already knew how to do that.

Sandage:

I knew how to do that from Illinois, and I knew how to use the 60-inch from Baade’s instruction.

Weart:

Oh, so this helped to introduce you to Hubble.

Sandage:

That's right. And on the basis of the performance on that project, he then asked if I would work for him during the summer of 1950. It was that sequence. So the M33 plates and M31 plates of the bright irregular variables were a project before I had ever gone to the 200-inch. That project had finished off, and I was sent up as the observer for him on the 200-inch. The 200-inch program was to extend the finding of Cepheid variables beyond the local group. One knows the history, that in 1924-29 he got Cepheids in the three members of the local group. Now, with the 200-inch, there was the opportunity to carry that out further, and the nearest galaxies were in the M81 group. So the program was to attempt to get plates, three times a month, of members of the M81 group. M81Y and 2403, IC 2574, 3077, and a few irregular galaxies. To do that, I was sent up for the first two nights of the dark run, came back for three days, was sent up for two nights in the middle of the dark run, came back, and was sent up again for two nights at the end of the dark run. So I went up three times in every 14-day interval. That was to get a long enough time base so that in a year or two, one would have enough time base to get the periods. Well, it didn't work out that way. Finally, the project was put on a longer time scale, and it took 17 years to finally got all of the plates for that extension beyond the local group. But Hubble would take the plates as they came back from Palomar, blink them, find the Cepheids in that run, and start a log. Logs of M81 and 2403 were begun during that period.

Weart:

I see. This was clearly one program that Hubble was starting. I wonder what were the general feelings of astronomers-Hubble and Baacle and so con—about how this great new telescope should be used? Who should be using it and what it should be used for?

Sandage:

Well, it was a staff telescope. There were no concepts of national facilities or guest investigators for the big telescope. Bowen had begun a project of guest investigators for the Mt. Wilson-Palomar observatories in general, and there were few people on the outside who used the 200-inch, but mostly it was a staff instrument, and there were twelve astronomers on the staff at that time.

Weart:

It seems to me just barely enough to keep the thing fully used.

Sandage:

Well, most astronomers thought they didn't get enough time. The internal pressure on time was enormous. Staff astronomers could expect to get 35 to 40 nights a year. Now it's unheard-of.

Weart:

Yes, it must sound attractive to you now.

Sandage:

Yes, that's right. So Hubble's programs had that amount of time. Baade's programs had that amount of time. The cosmological effort is spelled out in Hubble's manuscript which was published by the American Philosophical Society.

Weart:

Right. I wondered, was this sort of a general feelingy that this was the program for the telescope?

Sandage:

That was Hubble's program. But half the staff were spectroscopists, and the three other people in the nebular department were not in Hubble's group. Hubble was—well, all staff members were individuals. There were no observatory programs per se. So Hubble's cosmological program, he carried on independent of Minkowski and independent of Baade. Humason was with Hubble. The extension of the red shift distance relation, the expansion property, was an enormous activity, and Hubble and Humason were involved in that. Baacle was involved in the stellar content of galaxies, the population concept, which had come from his resolution of M31, and he was all involved in extending that and sharpening the criteria.

Weart:

I see. So there would be competition for dark time between these programs, and also with the spectroscopists.

Sandage:

Yes. That's right.

Weart:

How were these resolved, do you know? You weren’t on the committee at that time.

Sandage:

Not at all. But there were few enough astronomers, and half were light-time astronomers, interested in high-dispersion spectroscopy, chemical composition and radial velocities and so on. For the Coude aspect of the 200-inchy when the moon was upy there was no competition at all. So there were only three groups vying for dark time, at that time.

Weart:

Hubble and Humason as a group, and Baade, and?

Sandage:

And Minkowski. Now, Greenstein came into that a little bit later, with his requirement for dark skies to do his blue stars in the halo. But that was two or three years later.

Weart:

I see. I guess I should ask alsoy before we leave this business about what it was like to observe, what was the social scene like on Mt. Wilson, and also on Mt. Palomar then—eating dinner and that sort of thing'?

Sandage:

Well, there was great formality on Mt. Wilson. It was a tradition of the past, maintained still up into those times, that all astronomers had to wear coat and tie to dinner, and linen tablecloths, linen napkins—a moderately late 19th century, Victorian type of atmosphere. Or maybe 20th century of the 1920s, 1930s. Perhaps science was much more formal in those days. It was after the war that the informality really began to come. The graduate students came, the people coming back from the war began to be infused into the intellectual life of the universities, and it became more normal—normal's not the right word; much less formal.

Weart:

That happened fairly quickly then?

Sandage:

That happened, I guess, with about the second generation of graduate students. The first generation of graduate students still was operating under the social rules of the past. Then at the Athenaeum (faculty club) on campus, coats and ties were in order. There was no alcohol at the Athenaeum. There was a very strict social demeanor, and that infused itself on both mountains as well.

Weart:

I see. But on both mountains coat and tie were dropped fairly soon?

Sandage:

About 1955, I think when a large number of graduate students came and they just wouldn't stand for it. But the staff had then also started to leave on Mt. Wilson. Baade didn't observe on the mountain (Mt. Wilson), Minkowski didn't observe on the mountain. A group of younger people started to come in. We still have linen tablecloths and napkins on Mt. Wilson.

Weart:

What about on Palomar'?

Sandage:

No, it's much more informal.

Weart:

Always was more informal?

Sandage:

Yes, that's right, although there were periods when the dining was more formal.

Weart:

I see. Are both mountains still a place where you exchange ideas and talk much with other people?

Sandage:

Oh sure.

Weart:

You meet a lot of people there?

Sandage:

You meet a lot of people. A lot of graduate students come up, and it's a place where the lunchtime and dinnertime conversation is very animated. It all depends upon who's there, clearly, but it’s a place where the students normally have a chance to meet on a very informal basis with the staff.

Weart:

Has this always been true?

Sandage:

It's always been true.

Weart:

I see. Now, you became, in a sense, Baade's Ph.D. student.

Sandage:

That's right.

Weart:

How did that come about?

Sandage:

I wanted to do something in stellar astronomy. Galaxies, cosmology, stellar astronomy. And Hubble said he didn't have a problem.

Weart:

In that field.

Sandage:

That's right, for me to do as a Ph.D. Baade wanted students to do projects, and the location of the Main Sequence in globular clusters was needed. In Baade’s understanding of the populations at that time, it was fairly clear that the Main Sequence had to exist, but it had not been reached.

Weart:

Oh, that was understood?

Sandage:

Well, the H-R diagram of globular clusters was a mystery, in the sense that it was not like anything in the local neighborhood, and I don't know whether it was understood or not. It had to have been understood, though; I think people had to believe that the Main Sequence existed.

Weart:

This is one of the questions I wanted to ask you.

Sandage:

Yes. I would have to look back at the ASP^[7] articles at the dedication of the 200-inch. We can do that.

Weart:

OK. That's there, anyway. I admit I didn’t check that out to see.

Sandage:

I didn't either, and I guess I just implicity believed that the hindsight of everyone is such that the Main Sequence had to be there, and we were looking for it. When it came, it was not a surprise.

Weart:

It wasn't a surprise when you—

Sandage:

It was not a surprise when finally the Main Sequences of M3 and M92 were located. It was understood that one finally had gotten to a part of the H-R diagram that was understood.

Weart:

I see. It wasn't, "Hello, what's this?" It was, "Oh, there it is finally.''

Sandage:

Yes, "There it is finally." I think that it was the feeling. That's a good way to put it.

Weart:

What was Baade like? What were your relations with him like?

Sandage:

Well, he was a rather different man from Hubble. He was very gregarious, very outgoing, very interested in people. He told stories about people, and many times on himself, always had a joke. He read very widely, and he synthesized things from the literature. He would take pieces of information about spectral classification of Mira variables, for example, or high-velocity stars, and finally, when the time came, his mind was prepared to put all those pieces of information together. He devoured the literature.

Weart:

What were your personal relations with him? How much direction did he exert?

Sandage:

He didn't exert any direction, in the sense that the project was so well defined that after the plate material had been obtained, then it was clear what had to be done. Again, the training at Illinois and the observational training before the Ph.D. were so complete that the methods were quite well in hand. So in a sense, after the Ph.D. thesis project had been started, it was not a matter of training to extract the material. I think I saw Baade twice during the year and a half that I was doing the thesis work, to show him results, or to ask his opinion.

Weart:

I see. Were you seeing Hubble at the same time?

Sandage:

I was observing all this time on Hubble's program, on the 200-inch. I was not involved in the analysis or interpretation of the Hubble material. The Hubble part of the telescope was divided into two aspects, one which we talked about, finding Cepheidsp and the other, the classification problem.

Weart:

So you were taking photographs of—-

Sandage:

Of galaxies, that's right. And it's from that, that I learned the classification sequence. I talked often with Hubble, because he had to organize the next month's observing, and he had to tell mer from the last monthy what I'd been doing wrong, whether the plates could be improved. Out of that, by osmosis, the whole of the Hubble classification sequence was understood, and the programs he was hoping to do with the material. I learned a great deal. But it's not true that I didn’t talk astronomy with Baade. I talked astronomy with Baade, because we were together on the mountain, and I guess I did talk to him about my thesis. But there was very little direction, since it was such a simple thesis.

Weart:

I see. You might just as well be talking about some other astronomical problem.

Sandage:

That's right. I learned an enormous amount from both of those people—through the contacts at Santa Barbara Street, not formally on the campus.

Weart:

I see. This was the time when Baade had recognized he couldn’t resolve RR Lyrae stars in M31, and he’d come out of course with Population I and II. Was there any strain between Baade and Hubble because of this?

Sandage:

No, I don’t think there was any undue strain. I think that they did not communicate as colleagues very much, from the whole period of 1930 to the 1950s. They were not in each other’s offices daily But they certainly did communicate. A very interesting aspect of the whole history of the populations started in 1938 with the discovery at Harvard of the Sculptor and Fornax type dwarf galaxies. Both are reachable from Mt. Wilson, and Hubble and Baade collaborated in a very important paper, where they found RR Lyrae stars in Fornax. They also found regular globular clusters in Fornax. And that time, with the resolution of the globular clusters in Fornax, and the RR Lyraes being present through the whole body, and no blued stars, that the whole model of Population II was laid out before them. Baade once told me, “It was all there but I didn’t understand any of it. I didn’t understand any of it until I resolved M31 in 1943. And all of this was in the back of my mind, he said, “The mind was being prepared, and suddenly, it all came clear that the Sculptor and Fornax systems were the key–“ to the whole understanding of what was then his pure Population II. Now, that work on Sculptor and Fornax was done by Baade and Hubble together; there’s a paper in the ASP about it.

Weart:

There’s a lot of questions I could ask you about Hubble, but some of that’s covered in the Bert Shapiro interview. (Pause to discuss Shapiro interview). We’ll integrate parts of the Shapiro interview, not all of it. It’s a different kind of interview from what we’re doing now. Now, a little more about your education. Were there any important seminars, journal clubs, or other informal places that you met?

Sandage:

There was a weekly journal club, every Wednesday afternoon. Primarily, in the beginning, staff members talked about their results.

Weart:

This is staff members from Santa Barbara Street?

Sandage:

From Santa Barbara Street, and then physicists began to become interested in astrophysics. Greenstein and Fowler had coupled, and then Hoyle was here, so that by the second year of Greenstein’s department, there was beginning to be a great ferment on the campus.

Weart:

I see, you waw it appearing.

Sandage:

That’s right, and by the end of our graduate training, astronomy really was a very important aspect of Cal Tech. I think Greenstein did an absolutely remarkable job, starting from nothing. Now if you look at astronomy on the campus, it dominates the physics department. The old programs of the physics department have been phased out, and so much is connected in some way with astronomy. Well, that's a result of the child that was conceived in 1948. Part of that atmosphere was beginning to ferment, and there were colloquia, so that now the staff came down from here (Santa Barbara Street) every Wednesday. So we knew what was going on. And the interaction on the mountain caused information to be exchanged, so againy by just the process that all these things were in the air in Pasadena, the students got educated.

Weart:

I see. How important was the beginning of nucleosynthesis work to your interest in evolution'?

Sandage:

None at all. The interest in evolution came about solely from the astronomical side—first getting the color-magnitude diagram of the globular clusters. That, incidently, as you know from the literature, started out as a joint project with Baum and Arp. Arp and I wanted to do some pre-thesis research, and Daade suggested we understand the color-magnitude diagrams of globular clusters first.

Weart:

This was simply his suggestion? Understand it in what sense?

Sandage:

Do some work on it, because since the time when Shapley stopped his work at Mt. Wilson in 1918, there had been no modern color-magnitude diagrams. Starting then with photoelectric photometry to calibrate photographic plates, and then measuring the photographic plates, we could make a quantum jump in accuracy, because the photoelectric technique was fairly new at that time. Baum had joined the staff here at Santa Barbara Street. He was a Ph.D. physicist out of Cal Tech, and joined the staff to become a photometrist. It seemed natural that we three could combine efforts; he would get the sequences of globular clusters, and Arp I would measure the plates using them.

Weart:

Now, let me interrupt just a minute. In reading the (Mt. Wilson and Palomar Observatories) Directors' Reports, the thing appears in this fashion, which is compatible with what you've said: Baade and Hubble were particularly interested in using clusters to find the distance to Andromeda, as a step to the distance scale. And they hoped to find G-type dwarfs and use them to find the distance to clusters. It doesn't say to find the Main Sequence, but it says to find the G-type dwarfs. If you read the Directors' Report it seems part of a grand cosmological program. I wonder, is this the way it came to you at the time?

Sandage:

No. Those Directors' Reports are interesting, because one justifies the programs one's doing for all sorts of reasons. Come the 15th of May, one has to put something down for the director in a week. The globular clusters themselves, I guess from what you say, were thought to be possible steps to reconcile the problem of the distance scale, by using their integrated luminosity in M31. But it was known at that time from Cristie’s work here, William Cristie, that the luminosity function of globular clusters has a very large dispersion, that is, they go from minus 5 absolute magnitude to minus 11 absolute magnitude. So it's not a simple matter to use globular clusters per se to get the distance to Andromeda. But there was an outstanding problem with the globular clusters in M31. Hubble had discovered them, had measured them in M31, and the mean absolute magnitude of his sample was different from the mean absolute magnitude of the galactic globular clusters, even with Baade's revision scale. So that was a thorn, in the problem of whether Baade's revision was right or not. So I can understand what you've said, in the Directors' Report, that maybe one was looking for a magnitude and a half error in the calibration.

Weart:

That would be one reason for doing that.

Sandage:

That's right. Baade probably had all these things in mind, and we didn't see the big picture as graduate students. He just said, "Why don't you start to work on globular clusters?"

Weart:

I see. And you knew a bit about populations, so it was interesting from that side too.

Sandage:

Yes. It was then, in the beginning stages of trying to get the color-magnitude diagram of M92, on the bright side. Then Baade took Arp and I to the 60-inch on Mt. Wilson and gave us that training.

Weart:

I see. I find this interesting because for you it was an early large program; so before we get to interpretation, I'd like to ask you how this work was done. You measured over 1000 stars, photographically. Who did this and how was it done?

Sandage:

Well, there was the invention of a new type of photometer for photographic plates, by Siedentopf in Germany, and this was an iris photometer. The idea was taken up again after the war by Martin Schwarzchild, who likes to make gadgets—being a strong theoretician, that's interesting, but he does like to do things like that—he contacted an instrument maker by the name of (Lloyd) Eichner near Princeton, New Jersey. Eichner began to make iris photometers. Baade and Schwarzchild were very close friends, and we got on the campus a large Eichner iris photometer to measure 14-inch plates. This was the best measuring instrument for photographic plates then in existence in the United States. And it was on that machine that we started to measure these globular cluster plates. Arp and I did all the measurements for M92.

Weart:

You took turns?

Sandage:

That's right. And learned how to make the reductions, learned the idiosyncracies of the machine. It was on that machine that later my thesis problem on M3 was started. The M92 work was pre-thesis, and we did locate the Main Sequence, in a preliminary way the first year, and then in a more definitive way about five months later, with plates taken by Baade for us with the 200-inch.

Weart:

I see, and then Baum’s role was to help with the photometry, the photoelectric standards, I suppose.

Sandage:

That's right.

Weart:

I see. OK, now if you look at that paper...^[8] (gets copy of paper) I just wanted to show you what it looks like. It's really just barely the Main Sequence.

Sandage:

That's right. OKy this was from a symposium that was held, and the paper was given by Schwarzchild.

Weart:

Oh, he delivered it?

Sandage:

He delivered the paper. This was our first diagram, and the M92 Main Sequence was not improved from that for a long time.

Weart:

Nevertheless, even on the basis of that, you say, one sees the Main Sequence and the slope agrees with the regular Main Sequence, so clearly you were expecting—

Sandage:

Well, I think here the results of my thesis are also included, on M3.

Weart:

That's right.

Sandage:

This diagram that was published in AJ does not have the M3 diagram.

Weart:

I see, and the M3 one was a little clearer?

Sandage:

It was quite a bit clearer. That was the second cluster done, and it was then extended about two magnitudes beyond—-

Weart:

So you could very clearly see the slope—I see, I wondered about that. Did you have any connection in your mind at that point, before your work with Schwarzchild started, with theory, with evolution?

Sandage:

None whatsoever, nor did Baade. It was very clear that the use of the Main Sequence that Baade was interested in was for distance-scale problems, to calibrate the absolute magnitude of the RR Lyrae stars.

Weart:

I see. The fact that there weren't any beyond FO or whatever, that is, that the Main Sequence stoppedy was just something that was not understood?

Sandage:

It was not understood in California.

Weart:

I see. OK. Also, someone noticed the RR Lyrae gap.

Sandage:

Yes. That had been discovered by Martin Schwarzchild in 1935, at Agassiz Station with the Harvard 60-inch; he had shown that all of the variables in M3 were confined to that gap. No other work had been done since that timer and all we did was to show that the diagrams of M92 and M3 were similar to what he had found earlier.

Weart:

I see, it was simply a confirmation.

Sandage:

That's right.

Weart:

OK. Then how did it come about that you went to work with Schwarzchild?

Sandage:

Schwarzchild used to come out every year to Pasadena and work with the 100-inch and 60-inch on Mt. Wilson, and he was very good friends with Baade. One afternoon I came up from the basement here and had a diagram in my hand which was the Main Sequence of M3, the first results of the M3 work, and I was going to show this to Baacle. Baade and Schwarzchild were in the hall at the library entrance, and I showed it to Baade, and Baade showed it to Schwarzchild, and they became very excited at the Main Sequencers existence and where it turned off. But I don't remember at that time any comment about the explanation. Schwarzchild then, several days later, said, "Come back to Princeton when you finish with the observational work, and stay for a yeary back there." There was no connection, in my mind of the tracks off the Main Sequence that Opik had predicted and that no one took seriously at all. And I did not know of the Chandrasekhar-Schonberg limit, of some catastrophe happening. And Schwarzchild had not at that time done his gravitational collapsed core models, which then made clear what the next step beyond Chandrasekhar-Schonberg was. I didn't go back to Princeton for another year. In the meantime, Schwarzchild had begun the stop beyond the Schonberg-Chandrasekhar, with students, of whom Oke was the first. There is a fundamental paper, which really was the crucial paper, by Oke and Schwarzchild. It was all set up then, to carry the thing the next stage further, to add gravitational collapse to the models, which Schwarzchild did. I learned numerical integration techniques—this was the time before computers, or computers were just coming in.

Weart:

So when you went there, you already found Schwarzchild interested in gravitational contraction of the core?

Sandage:

That's right.

Weart:

And you already found him interested in comparing it with the diagram for globular clusters?

Sandage:

I don't know when that connection took place.

Weart:

Well, in this paper of Oke and Schwarzchild that you mentioned, I'm quoting, "It is tempting to compare (this) with the...diagram for globular clusters."^[9]

Sandage:

OK.

Weart:

They had a temptation, but they hadn't done it.

Sandage:

Somewhere in that period was when I first understood that what was happening in the core was connected with the evolution off the Main Sequence we were observing in globular cluster. Now, that was entirely a product Schwarzchild and his group. It was a Princeton situation, that connection between the observations and the moving off the Main Sequence. Schwarzchild really is the father of that idea.

Weart:

OK. I'm interested in another feature of this paper.^[10] So far as I know, it's the first paper where the individual models are determined successively, in the same sequence in which the star actually evolves, rather than just making a family of solutions and picking things out from them.

Sandage:

Well, I don't know whether it was the first or not, but clearly one could follow a given star through a series of configurations. Now, actually Chandrasekhar and Schonberg did the same thing. They had to follow what happened in the various zones to the entire structure of the star, as the burning continued.

Weart:

I'll have to check whether they just made a family, then picked successive ones. Anyway, it didn't strike you at the time as being an innovation?

Sandage:

No. But what did then strike everyone involved in this was that, given the notion that stars start on the Main Sequence and move off, then, regardless of the physics, just that statement in itself was the key to understanding of all of the observational properties of the H-R diagrams. Now Strömgren, in a very major way, had anticipated all of his, in a series of papers in ZEITSCHRIFT FUR ASTROPHYSIK and in the HARVARD BULLETINS. Kuiper had put together the Trumpler diagrams for galactic clusters, and Strömgren interpreted this in terms of varying hydrogen abundance—not in terms of evolution, of the change of structure of the star with time—but the interpretation of the H-R diagram as a sequence of some parameter varying had a long history. The next five years were ones of consolidation of the facts of observation concerning the H-R diagram, with just the simple statement that stars move off the main sequence rather than up that sequence.

Weart:

You immediately used this to produce globular cluster ages, in fact?

Sandage:

That's right, and they were all too small.

Weart:

How did you feel at the time about the accuracy, this three billion years?

Sandage:

I think neither Schwarzchild nor I thought that that was very accurate, the crucial important thing for us was the qualitative statement that you can get ages. We've joked privately many times about the statement we made in the paper, that it was fairly precise determination of the age—I think we gave 3.1.

Weart:

But then you said, "Within a factor of 2," or whatever.

Sandage:

And we completely neglected the time on the pretermination point of the Main Sequence; we didn't make the calculation correctly. We understood fairly soon why, even with the models we had, it was too small an age. But I think for us the important thing was the qualitative statement that ages could be determined by the turnoff point.

Weart:

What about the attempt to go on into the helium burning stage? Did you spend much effort on this?

Sandage:

I didn't at all. I would like to state that the merit of that paper was Schwarzchild's physical insight and mathematical abilityy and I was there as a post-doc student. He is an extraordinarily generous man to his students. He then went on later with Hoyle to understand in great detail the nature of the giant branch, the convective aspects of the Hayashi track, the placement of the track as a function of metal abundance, and the helium flash; in fact, the last 20 years of his work have been attempting to understand in the minutest detail all aspects of the globular cluster giants. What I did when I came back was to return to the observational aspects of the work and branch out into galactic clusters, and attempt then to use the age dating possibility to understand the various layerings that have been made in our galaxy.

Weart:

We'll get back to that. First, I'm just curious—this was the one big theoretical thing you did—how did you feel? What were you doing, turning a crank on a hand calculator?

Sandage:

Yes,.exactly right.

Weart:

Was Härm there doing that?

Sandage:

Yes.

Weart:

Was he helping on that?

Sandage:

Härm taught me how to solve four linear simultaneous differential equations numerically.

Weart:

I see. And then you did?

Sandage:

Yes. He and I became very good friends.

Weart:

What sort of role did he play in all this work?

Sandage:

He was Martin's computer expert. He did everything by hand when he first came, and then he learned the big machines. The Princeton experience was a very interesting one, because computers were just being invented by von Neumann at the time, and one of the very early computers was on the campus of the Institute for Advanced Study.

Weart:

Was that the MANIAC?

Sandage:

It was one that was paid for by the weather people. I don't know the name. Anyway, Schwarzchild would take Härm and me over to the Institute, and I met von Neumann at that time. The new computer was an amazing thing to watch; here's Johnny von Neumann with his child, like he always was, very excited about everything. But none of the actual numerical work on this paper, nor two or three subsequent papers of Schwarzchild and his students were done on the machine. They were all done by hand.

Weart:

I see. How was it to be doing theory rather than observational work?

Sandage:

I found it very difficult. Theory came a lot easier for me quite a bit later, when the cosmology aspect came; that type of theory was much easier. The physical insight took quite a while to come by, and I felt much more at home around telescopes and getting the data by observation. But it's clear that one really cannot do science without a fusion of both, and I think observational astronomy is sterile unless you know where you're going by some theoretical concepts.

Weart:

You had pretty much been oriented towards observation from the beginning?

Sandage:

That's right.

Weart:

Tell me, what was it like to be making a discovery, or to be in on a discovery? I don't know whether you'd say the real thing was there at Princeton or before when you found the Main Sequence—what was it like?

Sandage:

I can't really remember any moment of great insight at either of these times. The moment of discovery connected with this aspect came several years later, with the composite H-R diagram for open clusters—about a year after returning, I would say—when that all was understanclabley and the detailed luminosity functions; why there are two humps; why there are giants at zero absolute magnitude, G giants, G dwarfs; and why there is a gap between themy why there is a Hertzsprung gap; why everything funnels into absolute magnitude zero giants from Main Sequence points of plus 4 to zero; why there are no sub-giants later than K3. All these observational facts had been known from 1920, with the composite H-R diagram of open clustersp and just the statement that stars move off the Main Sequence, all of that fell into place.

Weart:

I see. How did that feel?

Sandage:

Oh, I thought that that aspect of observational astronomy could be used as a tool theny to really understand the formation and evolution of the galaxy. A certain sense of elation, but also a very strong sense that this was only the beginning, that having now understood the tool, that the tool really was so sharp, or rather not sharp but so powerful, that even more complicated and deeper problems were now open to attack.

Weart:

I see. Coming to this realization, did that make any difference in your feelings about things in general?

Sandage:

No, not really. The world had always been rational, in the sense that the equations of physics were always real, in the sense of Plato's archetype; the world is not irrational and I think the brain can understand the logic of the universe. But an elation, that something as un-understandable as the H-R diagram was really understandable by the laws of physics, by chemical of the stars, and just giving the stars enough time to age. But, nevertheless, it was a time of indescribable personal excitement that just got better and better with each year from 1950 to at least the 1960s. (Break for lunch)

Weart:

We're resuming after a lunch break. How did you get a job here at Santa Barbara Street? In the first place, had you wanted a job there?

Sandage:

Oh, sure. This was the goal of everything that I'd hoped to do from the time—from 1941 when I first visited Mt. Wilson.

Weart:

I suppose some of your coming up here to start working with Hubble and Baade–

Sandage:

Well, there was no conscious effort to sell myself, in that sense. And there was no indication that a job was available. But clearly this was the Mecca, this was the place that I wanted eventually to end up at. I had no idea that I'd be offered a job right out of graduate school. What happened was, the experience with Hubble on the M33 plates, the 100-inch and the 60-inch plates, was sufficiently goody and demonstrated at least the ability to transfer magnitudes and eye-estimate objects, so that Hubble apparently supported the belief to put me on the staff. It was unexpected. Bowen, after an observatory committee meeting down on the campus one day in 1952p before I went to Princeton, asked me whether I'd like a job or not. I hadnft gotten a Ph.D. yet, but I'd been observing at Palomar. Clearly a lot of people were retiring up here, and they were having to bring in young people to replenish their staff. It was a surprising event, because normally one expects to go out; this is the final placer after proving oneself on the outside. But I accepted Bowen's offer, and then went to Princeton for eight months without a Ph.D.—it was a postdoc but I didn't have a Ph.D. I joined the staff in 1952 and didn't have a Ph.D.; I hadn’t finished my thesis. So at the time that I worked with Schwarzchild, I had not written my thesis yet. I came back herey joined the staffy and was on the staff when 1: finished writing my thesis in 1953.

Weart:

I see. You had a fellowship at Princeton. Did Schwarzchild get that for you?

Sandage:

There was a fellowship called the Peyton Advanced Fellowship. Peyton was a good friend of the Princeton astronomy department, and that fellowship is a continuing thing; I had one of the early ones.

Weart:

I see. Was there any particular change you noticed when you came up to work here? Did it seem different to your from either being at Cal Tech or Princeton? The social atmosphere'?

Sandage:

Yes. This was the most famous observatory in the world, and just to come in the building was like entering Valhalla. Being quite withdrawn as a person in those early days, it was a very high-pressure responsibility. I felt that for 15 years, I think—to come into this place was a great privilege. I still feel it, but not in that same way.

Weart:

It put a good deal of pressure on you.

Sandage:

Oh, I put pressure on myself, to live up to that responsibility. But the first time, for example, in working with 100-inch or 200-inch plates—they were like the plates of Moses. I knew I had to be an astronomer when I was eleven, and everything culminated in my mind of coming here later in the career; but it didn't work I that way.

Weart:

Did this affect the type of work you were likely to do, the fact that you we here rather than being, say, at Princeton?

Sandage:

Oh, sure. This was an observational post. The tradition of Mt. Wilson was to collect the observations and make some inductive statements about what they meant. The interplay between theory and observation has always been important, but generally from southern California one induced from the observations up a level, instead of deduced down from the to I knew that it was and that observational opportunity. But it was clear from theory that one had to bring theory into all this, to make some sense out of it. The first years were a continuation of the globular cluster work, and the evolution that was then taught to me at Princeton, and trying to make something out of that.

Weart:

It started with the access to the telescope?

Sandage:

Oh, sure, the types of work that could be done, the very faint Main Sequence determination in globular clusters, the setting up of faint magnitude sequences in NGC 2403 to get the Cepheids—all that was unique to this place. No other place could, for example, at that timey have detected Cepheid variables whose maximum reached only 21 or 21-1/2 magnitude. There was a responsiblity to use the telescope. It was a continuation of Hubble's program, a strengthening of the foundations of the extragalactic distance scale, the recalibration, the carrying forward of Baade’s 1-1/2 magnitude revision inside the local group. Everyone both inside and outside the observatory, in the rest of the United States, really expected the 200-inch to do these sorts of tasks. By accepting that post, you already knew what was expected of you.

Weart:

I see. Now, to get down to the details of your work for the next decade, before the quasars for example, I think we ought to try to separate into, firsty the extragalactic work, and then come back to discuss the work on clusters and so forth, OK? Just to separate it.

Sandage:

Sure.

Weart:

You already mentioned the Hubble program, but I wonder, could you tell me about the growth of your interest in the Hubble diaqramy and in Ho and qo?

Sandage:

I think it's true to say that Hubble and Humason were certainly aware of the tremendous impact on the philosophy of science that the discovery that the expansion exists had. Hubble determined the time scale—at an early time you know that there was this problem of the discrepant time scale, it being much too short.

Weart:

Yes. You were aware of that when you were still a student at Cal Tech?

Sandage:

Yes. In fact, I think in THE REALM OF THE NEBULAE it states that 1.8 billion years is short compared to the age of the crust of the earth. And I think Schwarzchild and I knew the answer we wanted to get for the age of the globular clusters in 1952; we in essence wanted it to come out 3 x 109 (years), because that's what the geologists said the crust of the earth was. I'm rationalizing a bit, but that age at that time seemed quite logical.

Weart:

Anyway you didn't throw it out and go back to the drawing board.

Sandage:

No. And Baade’s revision of 1 1/2 magnitudes, for the moment, seemed to solve that particular problem. Baade never claimed that his revision of the distance scale meant anything except the distance to Andromeday and that's a very local step. it was the public press and the people outside the field that generalized, saying that the whole distance scale of the universe was increased by a factor of only two.

Weart:

Nevertheless, you say that at least for you it gave some hope that the age problem was on the way to a solution.

Sandage:

That's right. Clearly, the ability to age date the stars, and the parallel question of the distances to galaxies beyond the local group, were tied up in some dimly perceived way, in 1952-1960. It was not really as clear then as it is now what the predictions of the Friedman models were; that to fit the observations that there really are only two numbers that need to be determined, namely, the Hubble constant and the deceleration parameter. It wasn't clear at that time, then Hoyle began to give a series of lectures at Cal Tech. I think that series of lectures, plus talking with Robertson, who had the series expansion of qo (not in closed form) and then finally the 1958 paper by Mattig, which is a fundamental paper in ASTRONOMISCHE NACHRICHTEN, that the Friedman equations could be thrown into a closed form containing only the observables—that changed the whole direction (from what) the series expansion cosmologies had done up to that time. Heckmann, Robertson, and McVittie were the three principal theorecticians who continued to throw everything into series expansionsy and as the red shifts got bigger and bigger, clearly those series expansions were only approximations. And they didn't give the global picture anyway. So as these developments became clearer to me and with the closed form of Mattig, then for the first timey in my mind at least, it seemed possible that the observations could say something really globally. Now, Hubble had said that before. Robertson had said that before. But I hadn't understood it until perhaps 1958.

Weart:

Well, you were clearly interested before that. There's this 1956 paper of Humason, Mayall and SANDAGE^[11], where you wrote the sections on the Hubble diagram. You were interested in departures from linearity; there seemed to be some deceleration. Of course, that doesn't require one to do the mathematical analysis, it's just looking at the shape of the Hubble diagram. I'm curious, how did it come about that you wrote this section on the Hubble diagram in that?

Sandage:

Again, it was very fortunate to be at the right place at the right time. Humason and Mayall had begun this long program of systematically getting red shifts of the brighter galaxies. They started that in 1936. Humason took it to the limit of the 100 inch, and this was the prime program that he carried forward with the 20 inch. Mayall did the same at Lick. The combination of the red shifts and the apparent magnitudes was needed, and they were not interested in doing that themselves. I had begun, with Hubble, to get better magnitudes by modern methods of photoelectric photometry and schraffierkassette observations on the 200-inch, as part of the program Hubble started before he died. So logically it was an extension of the things that I had been brought into as Hubbleys assistanty and it seemed an easy thing for Humason and Mayall to get the last part of the analysis done by having it done that way. This enormous amount of work that they had done was combined, then, with a third section on the magnitudesy and that's where the three authors came in.

Weart:

You had measured the magnitudes?

Sandage:

Not for the field galaxies, only for the clusters. Edison Pettit had measured most of the magnitudes for the field galaxies through apertures of various sizes on Mt. Wilson. That was not the total magnitudes; one bad to correct those for aperature effect to an asymptotic value, so that took an analysis of how to go about doing that. That probably-took eight months or nine months for me to understand the problem, and then I had to measure the apparent diameters of all of the galaxies that Humason and Mayall had red shifts for and that Pettit had aperture measures for, and correct, those to some sort of total magnitude by the growth curves I had invented.

Weart:

I see. Humason had asked you to undertake this?

Sandage:

That's right, I was asked to do that.

Weart:

He just came around and said, "How would you like to do a Hubble diagram?

Sandage:

That's right.

Weart:

It must have been exciting for you.

Sandage:

Oh, it was tremendous. To be asked to be involved in what was at that time the most fundamental connection between the expansion of the universe and the observational data—that just doesn't happen very often.

Weart:

You were clearly interested in departures from linearity. Did this have to do with interest in steady-state theory?

Sandage:

No, I think really the steady-state theory was dead as soon as it was formulated. And almost every observational astronomer felt that way.That's because the arguments of Baade were so persuasive, that all galaxies are the same age—when he resolved the background disc of M31, and generalized that to the discs of other galaxies. And with the understanding of the agreement of the time scales, the expansion time scale and the age di-iAting of the stars, it was clear that steady-state really could riot be the case, because no young galaxies were known. So I think it's true to say that no one in Southern California ever took steady-state seriously.

Weart:

I see. What were your feelings about the Big Bang? How did it seem to you to be living in a universe that started with a Big Bang?

Sandage:

Well, having read REALM OF THE NEBULAE when I was a little boy, the initial shock was all over at that time. It was miraculous. But that's what the data indicated. And the agreement of the time scales, and the linearity? were crucial aspects. The fact that the relation was linear is most important, because that’s the only type of velocity field where a singularity is possible. And the fact that the observations indicate that, and the time scales now agreed, based on the globular cluster work and Baade’s correction–that the spectral lines are in fact shifted progressively, the greater the distance, is certainly one of the premier facts of all of science. But it's true. I mean—how does one feel in the presence of reality, you're asking. Well, the whole thing is such a miracle, in a sense. I still feel that it's like asking, what do you feel like when you realize that the charge on every electron is precisely the same? Well—it's an incredible miracle.

Weart:

This was something that happened to you when you were taking physics, I suppose.

Sandage:

The charge on the electron?

Weart:

yes.

Sandage:

I guess everyone interested in physics knew that from the time they opened their first book.

Weart:

It was not something that came to you suddenly.

Sandage:

I think that the whole world of physics is so other—worldly so un-understandable, at that level, such a miracle at that level, that you don't need science fiction to be excited. And the initerconnectivity underneath it all is so beautiful, that the world out there is—miraculous.

Weart:

By the way, did you ever read much science fiction?

Sandage:

Not very much, no. I found real science much more amazing

Weart:

OK, to get back to the paper: you noticed that there seemed to be a slight departure from linearity for the cluster data for which you had measured the magnitudes, and of course you said in the paper that since there wasn't enough known about stellar evolution, it didn't mean too much at that point. But I'm wondering, were you concerned at that point about whether the universe was open or closed? Was that an interest?

Sandage:

On yes, sure. At that time, all of the models were well understood from the work of Robertson and LL maitre and McVittie and Heckmann, and it was clear that the sign of the second order term, which one now calls qo (but we didn't call it that in 1956), determines the geometry. The whole philosophy of general relativity was understood, in a sense that, you give me the energy density or the dynamics or the mass, and I'll give you the intrinsic geometry (the space curvature).

Weart:

If you can think back to your feelings in the fifties, was there any way that you would have preferred the universe to turn out to be, open or closed?

Sandage:

Yes. For some reason I would like to have it closed; I would like to believe that you could see everything there was to see, eventually, that it was a closed entity. I don't know why that is. It's not a strong feeling.

Weart:

You felt the same then as you do now?

Sandage:

Well, now I really strongly believe that the data, not the expansion data but the other clatay indicate that there's not enough mass density to close it. So that regardless of how I'd like it, I think it's open.

Weart:

This is actually later, but I noticed in one of your papers, around 1975, you say, "Unfortunately, the universe is open."

Sandage:

Did I say that? I also worked out the equations for an oscillating universe, in the one stint of very simple theory I ever did. And those were so interesting, because if you take the light travel time into account, in an oscillating universe the nearby parts, after the universe started to collapse, will be observed under blue shift, while the more distant ones will still be observed under red shift, at the present epoch—because the light hasn't had time to reach us from the current epoch. I thought that was so interesting, and there was a neat diagram that was drawn, and I would have liked to have seen those equations applicable.

Weart:

Is there anything you find aesthetically more appealing about a universe that not only goes and comes back, but then bounces, so it continues to oscillate?

Sandage:

Oh, not really, no. You can speculate, if you're a Descartesian, and believe that given all the positions and velocitiest you can predict the future, then we'll all be here at the same time doing the same thing in the next oscillation—that would be neat, you know. But that's a science fiction hope and not really part of science. Because Descartesian physics doesn't work. I meant the uncertainty principle doesn't permit any of that.

Weart:

So, it would be a nice thing.

Sandage:

Yes. But to think the universe happened only once—that makes it even more mysterious. It's outside the realm of science, what happened before the first microsecond. Why it got itself into that, how it got itself into that state? But that's no more mysterious than noting the tremendous complication of the chemical balance of the human body. You cut yourself—and why is it the white corpuscles know exactly where to go to cleanse the wound? That’s a miracle. And I don't believe that's due to progressive selection of the fittest. It's just too fine a mechanism. I don't know what I'm saying now. I don’t know what the next sentence is.

Weart:

You're talking about your feelings about it. That's important. important.

Sandage:

I don't mean that points to the the existence of the God of the philosophers, whatever that means. Newton's laws are God, in a sense. But find it all so rational and so amazingly beautiful and so mysterious. You asked about deceleration and why that's important. I think everyone from 1935 on realized that to follow the expansion in time would permit you to find most of the parameters of the model. Even Hubble tried to read deceleration into his data; in his book in 1938, OBSERVATIONAL APPROACH TO COSMOLOGY, he discusses the deceleration problem.

Weart:

OK. We have to come back to some of these things when we talk about your later work. What about the Stebbins-Whitford effect? You mentioned it in this paper. Did you take this seriously? Were you concerned about it'?

Sandage:

I must say something about my philosophy of science, in the sense that I'm not an empiricist completely. Any result that seems strange enough so that it goes against the foundations that were set down earlier is suspect. There’s a theorem due to Bondi, that if there's a conflict between observations and theory, it's usually the observations that are wrong. When he first said that, it sounded heretical; but in fact most initial observations of a phenomenon are wrong. Herbig has a very interesting statement; if you found something exciting, wait a night and it will disappear. So, the Stebbins-Whitford effect being un-understandable, in terms of any known rates of evolution, many people took an attitude of, well, let's wait and see. That’s the same situation on a number of current things that look so strange. One has to know at each stage of an investigation which “facts” to ignore. It is here that intuition plays such a crucial role in the “road to success.”

Weart:

I see, that was the way you felt about it at the time?

Sandage:

Yes. I think that was put in the background, as a thing that couldn't be explained, but would be explained some time, and not to worry about it.

Weart:

OK. I also want to ask you about the Hubble ATLAS OF GALAXIES, which I guess is one of the things you spent a lot of time on.

Sandage:

Yes. That was done out of a sense of responsibility toward Hubble. He had been pointing toward an illustrated atlas before he died, and I more or less inherited that project, with the 200-inch plates that had been taken as the second part of the observing program for the classification studies. It was natural just to put those together and illustrate his sequence.

Weart:

The obligation was your personal feeling? Or was this something people around the observatory felt should be done?

Sandage:

No, it was my personal feeling. I was not given that responsibility or that task to do, but it was a thing that it seemed should be done.

Weart:

What was your feeling toward Hubble? What was your personal relation or feeling toward him?

Sandage:

Well, I liked him very much. It was always a very formal relation; he was a very formal man. We never talked personally about very many things. He was one of the people who taught me a particular style in science. He really was an intellectual that was dedicated to truth; there was no sham about it. Truth meant a great deal to him, and he would not understand dishonesty in science. Because it gets you nowhere, if you're really trying to understand something bigger than outside of yourself. These are simple statements that every scientist understands.

Weart:

I'm interested, because these are some of the same things you-were saying about Edwards.

Sandage:

Right. Well, I think that most of the teachers that had influence on me were very serious people.

Weart:

Was there anything in your early home life that gave you this sort of an orientation towards life or towards how one should work?

Sandage:

I don't think so. There was certainly a stress toward straightforwardness and honesty. But the scientific method, as a method to uncover something that's outside of you, that's bigger than you—I guess I understood that by reading the lives of great scientists. The period that is crucial to me is the last half of the 19th century. The attitudes of Maxwell, of Riemann, of Hardy, of all of the founders of the methods of modern science, the precision, the lack of subjectivity, came from that reading.

Weart:

This would be from MATHEMATICS FOR THE MILLION and this sort of thing?

Sandage:

I can't remember, but I must have read E.T. Bell's THE MEN OF MATHEMATICS. I read Ball's THE GREAT ASTRONOMERS. I read a lot of biography, where science really was the building of an absolute cathedral.

Weart:

I see. Well, to return to scientific worky I'm interested in your discovery that what Hubble thought were bright stars were really H II regions.^[12] How did that come about?

Sandage:

When the moon is up, you can't take blue plates, but you can take red plates. So, with the 200-inch there, and often being assigned the ends of the runs, when the moon was up—-

Weart:

Oh, when the moon started to rise or whatever.

Sandage:

That's right. Or waiting for it to set to get the plates for the Cepheids on the program that I went up for—then you took red plates, because the scattered light from the moon is blue. And so I took a lot of red plates for a lot of nearby galaxies, and found these knots were H II regions. That's interesting in another way. Hubble had to assume something in order to do his distance scale at all, and it's a miracle, it seems to me, what he did in the way of method; even though the identification of some of the objects were wrong, his method has a magnificence about it. He once told me, "You known, it goes full circle. When I said in the paper that the absolute magnitudes of the brightest stars in these galaxies are -6.1, the spectroscopists on the Mt. Wilson staff said, 'You're crazy. There are no stars brighter than -3.'' Whereas in fact at -6 he was much too faint. The brightest stars are -9, and the brightest H II regions are -11. So he thought he was erring on the other side, of being too bright for his indications.

Weart:

That's interesting. Well, this H II regions business of course was only part of the further extension of the Hubble time—-where you go from 3 to 5 to around 13 billion years, which was done at that time. Of course this is still with a large uncertainty which everybody acknowledged. I'm interested in all this, and whether there may have been any role of discussions with people like Fowler, Burbridge, Hoyle on nucleosynthesis and age problems from that side?

Sandage:

No, not directly. But it's true that during this ten-year period there was great ferment, both on the Cal Tech campus and up here, and there were lots of interactions. The work that was being done in nucleosynthesis was clear to everyoney I mean? that it was going on. And the Burbidges were, and still are, very good friends, so that we would see each other every week; I would have dinner with the Burbidges often, and we discussed many things. But the actual fact of nuclear furnaces inside stars, and their role in evolution, is not important in the determination of the time scales nor the Hubble constant per se.

Weart:

OK. One of the things that was important, of course, was the Capheids. You've mentioned a couple of times your work on this. I'm interested in how the growth of your concern about this came about, the general history of your understanding of the Cepheids—up to this paper where you talk about the H II regions? where you have the discussion with Arp that the period-luminosity relation has a real width. What led up to that''..''

Sandage:

That's a natural outgrowth of the statement that stars move off the Main Sequence and go into the giant region in their evolution. It was well known, from all the studies in globular clusters that the RR Lyrae star gap had a width, and that width was appreciable. So, just from the most general principles of the period times the square root of the density being constant for any pulsating system under gravity, you could empirically derive the period-luminosity relation. You needn't put the physics in to understand what sustained the oscillation; if you gave yourself the fact that the stars did pulsate, you could predict that there would be a period-luminosity relationy by knowing how the radius and the density changes go in the H-R diagram. It turns out that Shapley and Henry Norris Russell had discussed this in a paper, and came to a conclusion of finite width. I didn't know that until much later. In terms of what I then knew about the finite width of the RR Lyrae star gap,—-

Weart:

—I see, that already gave you the feeling—

Sandage:

That's right—of an instability region. And then it was a simple step to go from that concept of a region to ask, suppose that region has finite width, what does that do to the periodluminosity relation? Then it was just a matter of cranking out. That was one of the very exciting times, I think, of science for me—the working out of the possibility that we would understand an intrinsic width to the period-luminosity relationy that there was a third parameter besides period and luminosity.

Weart:

Already looking for the blueness, in fact.

Sandage:

That's right. And that theme, then, pervaded the next ten years of the recalibration of the period-luminosity relation.

Weart:

What role did Arp play in this?

Sandage:

In the indication of the finite width, I think, nothing. He discussed, in terms of the predictions of the theoryp his observations in the Large and Small Magellanic Clouds that held gotten iin 1958 and 1959. The idea of a third parameter was new to me at that time, but as I say, looking back in the literature, Shapley and Russell discussed it in terms of a finite widthy in terms of spectral class—not in terms of color and density.

Weart:

I understand. Well, another thing that struck my eye—I’m going from paper to paper here, the only way I can structure it (laughter)... One thing that struck me is a well-known paper of yours, a review of the ability of the 200-inch telescope to discriminate between world models^[13]—you mention a "renewed interest in the cosmological problem." This is 1961. Was this something that you noticed, could you look about and see that there were more people interested in it than before?

Sandage:

Well, I think that I meant by that)a hope for the future, with the discovery by Mattig of closed forms for the equations that the observations could be fitted into. Not only that, but the existence of the 200-inch, being much more powerful than the 100-inch. I suppose I meant by that, renewed interest in the sense that now it looked possible to solve the problem.

Weart:

You felt back in the fifties, that very few people were interested in tht sort of work? Or just that they couldn't do it'?

Sandage:

I think very few people were actually interested. The 200-inch, again, was the only major telescope that could be used. I must say that I didn't work on cosmology myself for the first ten years of my career, I think because I didn't feel that I knew what to do in a meaningful way. The question of stellar evolution was much more important at that time, and looked like it was just falling out, and indeed it was. I think already one knew that one had to understand the stellar content of galaxies in look-back time, before any very distant red shift objects could be interpreted. It's true that Humason continued to try to push the observations of large red shifts, and I was involved in discovering clusters of galaxies for him to work on, by various techniques, with the restricted field of the 200-inch. But I think that except for the theoreticians, Hoyle with the steady-state and Robertson in attempting to quantify the models even more, there was not much observational cosmology.

Weart:

What about this particular paper? You came up with a qo of one, more or less—things were uncertain and there was a lot of discussion. What reaction did the community have to this paper?

Sandage:

When was that?

Weart:

That was the 1961 paper on the ability of the 200-inch telescope to discriminate between world models.

Sandage:

Well, in the next volume of the APJ there is a paper on the importance of the evolutionary correction.^[14] And there, the formal value of qo is one, but putting in what was known about the evolution up the M67-NGC 188 sequence, qo was claimed to be 0.2. That is, the formal value of lt with the evolutionary correction, went down.

Weart:

I'm curious about what sort of reaction you got.

Sandage:

I don't think there was any reaction at all. I think that interest in observation of cosmology, still was very small. There were very few people working in the field.

Weart:

So at meetings or whatever, you wouldn't get a big rush of interest in papers that you'd give on the subject?

Sandage:

I think it was felt by everyone involved that although these are interesting speculationsp the answer can only come once the red shift is pushed beyond two tenths or four tenths—only in the realm of red shifts in the order of one do the departures get really many sigma above the instrinsic scatter.

Weart:

I see. So it was just a matter of waiting.

Sandage:

And it looked impossible to get red shifts much larger than Minkowskils, of 0.46.

Weart:

With what was available at the time.

Sandage:

That's right, with ordninary nonlinear photographic detectors. The television subtractive detectors had not been invented, and the night sky barrier seemed crucial. But there was quite general interest in what was called "experimental geometry." You know, Gauss was the first to really take seriously the non-Euclidian aspects of geometry, by making measurements of the parallax. Hubble took that on, and it looked as if it could be really taken in principle, if we could get red shifts that large. So there was interest in that general problem, but everyone believed the observations were so poor that it proved nothing.

Weart:

I see. What about the growth of your interest in the evolution of luminosities? Did the Stebbins-Whitford effect stimulate interest in this, do you suppose?

Sandage:

I don't think the Stebbins-Whitford effect did, because the extremely rapid gradient of color with red shifts was believed impossible under any theory and therefore that the Stebbins-Whitford “effect” could not be real.

Weart:

So luminosity evolution was simply a basic thing that you always knew had to be figured out.

Sandage:

That’s right. And that came naturally, because one knew to understand the stellar content in terms of the H-R diagram, at that time, based on the earlier globular cluster work. The whole mechanism, the whole machinery for computing the evolutionary correction, was understood in 1960. You just looked, in the look-back time, how far the main sequence has burned down, and you know the luminosity relation along the giant branch, so you just sum those.

Weart:

Then let's get back to your work on thatr on clusters and evolution. First, I should ask, did you feel a close connection? Did you feel that you were doing evolution, eventually to be in service to cosmology, or was it simply an interesting thing in itself?

Sandage:

Both. In the back of my mind, as I remember it, I always knew that I would eventually go into cosmology, that is, turn the observational effort more directly into cosmology. It wasn't as ify in this period between 1952 and 1960, there was no cosmology, because Humason was going up to Palomar all the time, getting red shifts, and I was very close to Humason.

Weart:

Yes, and you did have a few papers.

Sandage:

But it was clear to me that evolution was much simpler, and it was falling out, and it had to be solved before one could get a handle on the other.

Weart:

OK. Let's go back then to the galactic star clusters. I'm particularly interested in M67, M3, M92 and so forth, and putting them together to make the combined diagram, the quest for the zero age main sequence stars. I'm also curious how it came about that it was found at this point?

Sandage:

That the age zero main sequence was found at that point?

Weart:

Yes, right. How you went about finding it.

Sandage:

Well, I was very close friends with Harold Johnson, along with Olin Eggen, who also is a very good friend; thoe two astronomers were involved in getting color-magnitude diagrams of galactic star clusters by photoelectric means. And it was clear from the Chandrasekhar-Schonberg evolution that near the turnoff point, the slopes of the main sequence had to be due to evolution. So it was clear that distances, determined by photometric means, had to be corrected for evolution. In a converstion with Harold Johnson, I can't reconstruct where, we spent two days talking about how to get photometric parallaxes. It was out of that conversation that came, at least for both of us, the statement that if one goes far enough down on the color-magnitude diagram, three to five magnitudes below the turnoff point, you will get to a point where the evolution is negligible. And if you have enough clusters, you can stack them and throw away the upper parts; and if there are no chemical differences, you can derive that way the age-zero main sequence.^[15]

Weart:

I see. So it was tied in at the beginning with concern about the distances.

Sandage:

That's right.

Weart:

Being able to compare different clusters, in a sense.

Sandage:

That's right. The distances themselves to galactic clusters were interesting, and I don't know whether the motivation to do that at that moment was because Cepheids were in clusters, and we knew we had to recalibrate the Cepheids period-luminosity relation, or whether that concern came later. It was 1957, this development by Johnson and myself, and we published separate papers on that.

Weart:

I'm interested in your work with collaborators, not only with Johnson but in general, how does that work?

Sandage:

Oh, very well. (laughter)

Weart:

But what's the mechanism of it? You've published a number of papers in collaboration.

Sandage:

I think collaboration works well only when there's mutual benefits on both sides. And in general I collaborate either with students that can help with the work, or with people that know so much more about a given aspect of a subject than I do, that we can both bring something to that. It's simple enough to say that it's easier to collaborate with some people than with others, but the secret of collaboration is to put yourself completely in the other person's hands and be flexible. Also, not to collaborate in detail but to take a certain segment of the thing and do it, and then combine at a later time.

Weart:

I see. Is there a particular role that you usually play?

Sandage:

It depends on whether I'm collaborating with students or I'm collaborating with a senior person. With students, I normally ask them to write a first draft. That's educational for them and it also gets the work done. Then I write the second draft. With people like Lynden-Bell and Eggen, where we did this collapsing galaxy paper, that brought so many diverse aspects together that none of the three of us could have written the paper by ourselves, and each author was responsible for a segment. Then we all put it together, in sessions, into a whole.

Weart:

I see. I'll have to ask you that when we get up to your paper. To continue with the Johnson and SANDAGE work on clusters, I was interested because at that point you got into things like the ultraviolet excess, leading into low metal abundance, and also at the same time you were studying blanketing of subdwarfs with Eggen, as you mentioned, and the Burbidges. I'm curious as to how much your interest was attracted by main-line astronomy, spectroscopy, toward the study of stars as individuals.

Sandage:

Very much so. I tend to be more interested in the duller aspects of astronomy, in the sense of those aspects which are calibration problems, as a calibration problem itself. We knew, for example, that along the road someplace we had to know the absolute magnitude of the RR Lyrae stars. Well, the very best way to get that is via the Main Sequences of globular clusters. But the idea in 19461 in a symposium which outlined the program for the 200-inch, of using the Main Sequence and just force-fitting to the Hyades was found not to work, because of the ultraviolet excess. So that the whole problem of main sequence fitting, which was in the back of everybody's mind, wouldn't work until the problem of blanketing, and whether in fact, there is a real subdwarf sequence, was solved. So that meant a side excursion had to be made, via the trig(onometric) parallax stars, of different ultraviolet excesses, and there's a paper with Eggen^[16], where we divide all of the tr arallax stars, of high weight into four ultraviolet excess groups, and see the position of the apparent main sequences. And that had to come, before we could use the main sequences of globular clusters. So there's a reason for all of these diversions. It was all funneled toward trying to get the absolute magnitude of the RR Lyrae stars, or the distance to globular clusters.

Weart:

OK, I guess that makes that pretty clear. I suppose it happened at the same time, this funneling of things into the red giant region, and a general understanding of the H-R diagram. I wonder if you could talk a little about that?

Sandage:

Well, that's the diagram right there (pointing at a combined diagram behind desk.

Weart:

Oh, is that the original one?

Sandage:

That's the original one. It's been hanging there for a long time.

Weart:

I guess you added a few things to it from time to time.

Sandage:

The position of the globular clusters was not in there at that time, because the question of the blanketing had not been solved. But the height of the giant branch of, say, the resolved stars in the disc of M31 and M33, the things Baade had resolved-it's crucial to understand what those are. Are those globular cluster stars? Or are they the more normal absolute magnitude zero stars that funnel into that region normally? You don't know that until you make a census of the color-magnitude diagrams of clusters of different ages. The whole topology of the H-R diagram and its explanation in terms of evolution was crucial to an understanding of what types of stars are resolved in the nearby galaxies. And that in turn was crucial to understand the sequence of events of laying down the disc and the halo of the nearby members of the local group, and then taking that out further.

Weart:

So you were quite interested in the overall problem of how galaxies evolve. In a sense this was in service of the picture of galactic evolution?

Sandage:

That's right. That, in combination with the properties of high-velocity stars—that is, the fact that the stars with the highest ellipticities of their orbits around the center (of the galaxy) are the ones that in our neighborhood seem to have the highest space motions, the correlation of that with metal abundance, and with age—

Weart:

Were you interested in that before your work with Eggen and Lynden-Bell started?

Sandage:

Oh yes. Because that seemed to hold the clue to some fundamental sequence of events in the early history of the laying down of our galaxy. It seemed that it would be impossible to understand how our galaxy formed from everything up until Baade's separation of the populations. And then the correlation of metallicity with the ellipticity of the orbits, and the nature of the halo versus disc of our galaxy—it seemed none of this would make sense. But with all of these separate isolated facts, it seemed possible that a scenario could be reached. Donald (Lynden-Bell) did all the dynamics, and Olin (Eggen) and I correlated the ultraviolet excesses and the ellipticities, and came up with this picture, which at least explained all the known facts at that time.^[17]

Weart:

How did it happen that you decided to do all this, and that the three of you come to work together on this?

Sandage:

Well, Eggen had been here for three or four years, and Lynden-Bell came over on a one or two year fellowship, and the ferment on stellar evolution was strong on the campus. The discovery of the ultraviolet excess, and the fact that stars with larger delta (U-B)'s had higher space motions—

Weart:

By whom?

Sandage:

Nancy Roman did that.

Weart:

That made it logical to try to make whole correlation.

Sandage:

Not only that, but then to look individual stars in the literature that had high velocity, an get new observations to see if the correlation was general, to s if the highest velocity stars had the highest ultraviolet excess which meant the lowest metal abundance. As those observations came in, it looked like the whole picture was just going to fall right out. We went to Lynden-Bell.

Weart:

You and Eggen?

Sandage:

That's right. We first had the idea that the pre-galaxy was in a state of hydrostatic equilibrium, before something happened. That the halo was held distended against its own gravity in its gaseous form.

Weart:

By random motions?

Sandage:

That's right. Then the stars formed by some process, and the collapse toward a disc took place. We asked Donald to look into the dynamics. Well, it soon was evident to Donald that the pre-galactic mass could not have been in a state of balance, that the temperature required was so high that no star formation could take place. The correlation still existed between the eccentricity and age of the stars, and that clearly must have meant, since the eccentricity cannot change with time (Donald proved that from the adiabatic invariance of the parameters of the orbits, but that seemed intuitively clear anyway)—from that, the whole picture of the Jeans instability collapse, in a continuous way, making these correlations possible, just fell out. The collapse took place by a separation of the protography mass from the expanding universe. This general picture had been present from observational data that had been coming in for four, five, or - six years, on the metallicity versus kinematics versus globular cluster main sequences versus blanketing.

Weart:

Even so, was it a surprise to you to find this sudden collapse of the galaxy?

Sandage:

The time scale of the collapse, which was forced on us by the analysis of the data, the free-fall collapse, was a surprise, yes.

Weart:

Did you feel you'd really got something?

Sandage:

I think all three of us felt good after that paper; that that paper was wrong in detail, but was a step.

Weart:

I see. Did this have any connection with your work, which I guess was starting around that time, on M82 filaments?

Sandage:

No, that was a complete side-issue That came about from the discovery of Roger Lynds of the kinematic field of M82; Roger urged me to take direct plates. So that was a result of a collaboration being asked for by Lynds, and most of that work in fact was due to Lynds.

Weart:

I see. Was there any idea at the time, when you studied the galaxy collapsing, that it was what we now call a violent collapse? Of did that sort of feeling come only after the quasars?

Sandage:

We never connected the general collapse formation picture of our galaxy with any of the things that came later.

Weart:

Yes. I'm just talking about a general feeling that the galaxy collapsing indicated that the early phase was somehow much more violent.

Sandage:

No. Hoyle had written a paper on general fragmentation in astronomy, about 1950, which had a great deal of influence on the thoughts that Jeans instabilities took place on all different scale lengths, and stars, globular clusters, associations, whole galaxies, were part of that spectrum.

Weart:

I see. You mentioned Hoyle, and I'm interested—you had a a paper, I guess in 1962, on comparison of observations of a number of clusters with Hoyle's models.^[18] I wonder, what sort of interactions did you have with Hoyle? He was here every year at that time?

Sandage:

Pretty much, yes. He had produced the most explicit and only published stel ar interior models that were detailed enough, in terms of radius and output luminosity of stars as they evolved, to permit comparison with the observations. So, if one were to age data in a more precise way instead of just quantitatively, one had to use models, and Hoyle's models were the only ones available.

Weart:

What kind of relations did you have with him personally?

Sandage:

Oh, we were colleagues. I sat in on cosmology courses of his; he was here every summer, and we talked very often.

Weart:

How was Hoyle regarded by people on Santa Barbara Street?

Sandage:

Hubble had essentially died off before Hoyle's heyday at Cal Tech, and Hubble maintained good relations with almost everyone, but the difference in philosophy between Hoyle and Hubble was profound. Hoyle, I think, was and is more interested in all the worlds that could be, instead of the world that is. Anything that is logically accepted within the realm of physics, even though you have to change the laws a little bit, is of interest for him to speculate on, whereas Hubble was an absolute empiricist, asking, "What is it really like?"

Weart:

Of course that was the tradition here also.

Sandage:

That's right, so the way Hoyle approaches science is really quite different from the way that Santa Barbara Street traditionally approached science—an attempt not to speculate, to take what the universe gives you. And that difference still holds today.

Weart:

Santa Barbara Street has changed a bit, I guess.

Sandage:

No, I don't think so. we're still very conservative—no, some of us are still very conservative.

Weart:

Most of you?

Sandage:

At least a few of us. (laughter)

Weart:

Did you have connections with other theorists, for example Schwarzchild? Were you discussing this work that was going on in terms of computations of stellar evolution? Did you get involved in that?

Sandage:

After I left Princeton, there was little interaction with the Eastern theorists in stellar evolution. But there was interaction with H.P. Robertson in cosmology until he died. Schwarzchild still came out here. He was influential in the series of papers on the semi-empirical evolutionary tracks. Without a detailed knowledge of the models you can, given the H-R diagram that's observed, and the luminosity function, the number of stars in each interval, reconstruct everything about the way the tracks have to go.

Weart:

This is very much your viewpoint, your approach to the problem, rather than the computer side of it?

Sandage:

That's right. That's strictly from the observations, saying how the tracks have to be in order to reproduce the observations. Now, that series of papers was very much influenced by talking with Schwarzchild.

Weart:

Because of course, you have to agree in the end.

Sandage:

Yes, that's right. And he also showed where I'd made mistakes in the initial formulation of the problem.

Weart:

I see, he was interested enough in it. Another thing around this period is NGC 188, also M67—finding them to be very old.^[19] In fact at that time it looked older than the Hubble age as then known. I'm interested in your development of this work, and also your interest in the chemical evolution pattern of the galaxy. We talked about that a little, I guess. You said that the fact that they are very old yet metal-rich indicated that nucleosynthesis in the galactic disc proceeded very rapidly within the early phase. Did this feed into your work with Eggen and Lynden-Bell?

Sandage:

Yes. The fact that the enrichment had been completed by the time that the disc was laid down was consistent with the model that was produced.

Weart:

What was the sequence of things there?

Sandage:

First of all, it was a shock that things so old were so metal-rich. You asked whether there was any input from the nucleosynthesis group into these ideas, and the answer was yes; the idea from the Fowler-Hoyle-Burbidge group was that there was gradual enrichment of the interstellar medium.

Weart:

It's that sort of basic idea—not the details, but the basic idea.

Sandage:

That's right. And then, if one could plot, as a function of age, the heavy element content, one expected to get a smooth, slowly rising curve. But in fact now, the oldest things in the disc, and the disc itself, being proved to be metal-rich, observationally, that 188 being as old as it was but as metal-rich, meant that it was not gradual enrichment, within the time that it took to lay the disc down but “catastrophically” rapid enrichment mean the beginning of disks formation.

Weart:

What happened—you found that first, but then, you found the collapse from another angle, but this helped to confirm it?

Sandage:

I think the two were really unrelated. They can be understood in terms of one another, but the development of one didn't have any influence on the development of the other idea.

Weart:

I see. It was only afterwards that you put them together.

Sandage:

That's right. In the paper on 188, thre is a sentence in italics which says, the metallicity of the disc had reached its present point at the time of its formation. That was in italics because that was so foreign, _Wt-t-ffe time—and a direct observational result—from the gradual enrichment due to nucleosynthesis. But it really had nothing to do with the Eggen, Lynden-Bell paper.

Weart:

I see, at the time it just seemed unexplainable.

Sandage:

That's right.

Weart:

I guess it made it a lot easier for you to imagine this rapid collapse when that time came around?

Sandage:

But out of that then came the idea—by the nucleosynthesis people on the campus—how can you manufacture so many heavy elements in the short time that's now required?

Weart:

I see, they would come around and you would be discussing this type of thing?

Sandage:

Yes. There was great interaction between here and the campus and everyplace else. It was a time of extraordinary ferment—a very, very charged atmosphere.

Weart:

More so than now?

Sandage:

Oh, I think it's very relaxed now. That may be only because we're more compartmentalized and don't see each other; or I don't see many people on the outside. But at the same time there was an absolute electric atmosphere, because for the first time in 30 or 40 years, the whole of this mysterious H-R diagram was being understood in terms of fundamental processes.

Weart:

And a lot of different things were being put together.

Sandage:

A lot of different things were coming together. Out of the rapid enrichment requirement came the idea that maybe all the metals were made in supermassive objects, that there were 106 solar masses and went through their evoi~tion in 103 years. During the collapse, you had 105 generations of these things. The supermassive bodies of Hoyle and Fowler were invented for this problem, but more specifically, later on, for the quasars.

Weart:

It was a very busy period and ask you about each of your papers; there were more on clusters, the initial luminosity function, and so forth. I am curious to ask some questions in general about how you work. How did you find the time? What was the rhythm of your work?

Sandage:

I look back on it as a time that was not any busier than now. I think that the frequency of finishing projects was greater then; the urgency and the pressure was greater, internally, than now. It just seemed like everything was so clear and laid out; after the observations were obtained, one just sat down and wrote the results up. I observed a lot, then like now. I expect I had 60 nights a year on the two mountains (Mt. Wilson and Palomar) on three of the world’s largest telescopes (60-inch, 100-inch, and 200-inch”).

Weart:

That's a lot of your time.

Sandage:

That's right. Last year, however, I went up for 105 nights.

Weart:

On the 100 or the 200-inch?

Sandage:

And Chile, and the 48-inch. The time on the mountain keeps a constraint on to finish things up the week or two you have in the valley. The pressure of knowing you're leaving the office, and having to change what you're doing, immediately causes you to try and finish things up.

Weart:

How does it work? Let's talk about both the (fifties and now) you spend a week here, then a few days on the mountain, then a week here?

Sandage:

That's right. At that time, I was still measuring most of my plates myself. That stopped about 1955, when we did get technical assistants here.

Weart:

I see. Would you be working on several studies during the same week, or did you do things sequentially?

Sandage:

There were five or six projects going on at the same time.

Weart:

Not only in terms of observating, also in terms of what plates you might be measuring?

Sandage:

That's right. And there were four or five observing programs going on, and you would tailor them to the nature of the conditions every night.

Weart:

You would have several lists, and that particular night, you would observe this or that, depending on the seeing or whatever.

Sandage:

That's right, if there were high clouds, you would not do photoelectric photometry but take direct plates, for example. There were usually two foci of the telescope with auxiliary instruments set up, and you could change from one to another.

Weart:

I see—if it got cloudy, or the moon came up, or whatever.

Sandage:

That's right. I had a Coude project for high-velocity stars for about five years. We were anxious, Eggen and I, to locate more high-velocity subdwarfs that were fainter, and fill out the bottom of our diagram, to see the correlation of metallicity with eccentricity of the orbit. out of that came about 150 new space motions for subdwarfs.

Weart:

Would there be sharing of plates, in the sense that you would be takng plates that would be mainly of interest to somebody else?

Sandage:

That happened quite rarely.

Weart:

You would just have several programs, or several programs cooperation with somebody else.

Sandage:

Yes. Normally there would be a collaborative effort on certain aspects that had been agreed on before.

Weart:

I see. Now, at that time, since you were a staff member, you would automatically get a certain number of nights, but even so, and particularly more recently, you have to put in a proposal for a program.

Sandage:

We've always had to do that.

Weart:

And it becomes more and more important as pressures on time get greater. But if you put in a proposal to spend so many nights observing such and such, but in fact, actually you have four programs—how does that work out?

Sandage:

Oh, you're very honest in the proposals and put down, "I'm working on eight projects."

Weart:

I see, so you line up all your projects, and you just want so much total telescope time for all of them.

Sandage:

That's right.

Weart:

Do they ever say, "Never mind this project, we're more interested in that one."?

Sandage:

Yes. But there's tremendous freedom at the observatory, in the sense that there's no observatory programs. You're judged, on the basis of past performance, whether good use is made of the time or not.

Weart:

What about an outside observer, one of the visitors? Do they also give a list of what they want to do?

Sandage:

That's right. And that raises an interesting problem, because clearly, there are fast-breaking areas, and the visitor, when he has the 200-inch and something breaks, is he permitted to change and do that? Well, there have been instances where that's happened, and that's caused some problems.

Weart:

I see, criticism occurs if they went to something that's not in their original program.

Sandage:

Oh, not here and there, but if it's a general attribute of the visitor to propose something, and then almost always do something else–

Weart:

I see, they may not be invited back.

Sandage:

Especially if he knows that a staff member is working on that. In astronomy, I think every research person knows what everyone else is working on. It's a very open field, and there's a lot of communication, and a lot of annual reports published—I'm not sure in physics, whether there's as many observatory reports per year-

Weart:

Oh, there's nothing like that, except in high-energy physics. High-energy physics people have some idea what's going on.

Sandage:

And there, the experiments are so complicated that you do just what you're-

Weart:

Oh, it takes three years to get yourself set up. But as an astronomer, it would be very easy—I suppose people have poached, so to speak?

Sandage:

Well, astronomers have a reputation, almost every astronomer now listed in the AAS book (membership directory), other astronomers know almost everything about them.

Weart:

I see. But of course, this is something you only learn as they come in.

Sandage:

Sure, that's right.

Weart:

So what happens? Are they warned or they won't be invited back or—?

Sandage:

It's a question of how strongly the committee feels. There are some very eager astronomers that publish a lot of papers that are still invited back, because even though they do other things and encroach upon the staff, their science is good, and one just lives with that competition.

Weart:

That's very interesting. I want to ask you more about the division of time later. But the, next thing I have down here is that you married Mary Connelley in 1959.

Sandage:

That's right.

Weart:

What was your wife's background and education when you met?

Sandage:

I gave a series of lectures at Harvard, in 1957, on stellar evolution, and she was an astronomy graduate student at Harvard. She was raised in Indiana, went to Indiana University in Bloomington, and went to Radcliffe in the graduate program. She then went to Mount Holyoke and took over the astronomy department there from Alice Farnsworth for three years, and we then got married in 1959.

Weart:

Has she maintained any separate career after that?

Sandage:

She did not in a strong way. She has edited Volume 9 of the Kuiper compendium, and that took a great deal of time. But she was the principal raiser of two children.

Weart:

That takes some time.

Sandage:

That took an enormous amount of time, and she liked to do that, and she has done a terrific job.

Weart:

In some of your papers you thanked her for help, discussions and so forth. How much help has she given?

Sandage:

In the early days, very much. In the big paper on the ability of the 200-inch to discriminate among world models, she checked all the integrations, and in fact did a few integrations that I couldn't do. Then she found that it was very hard to stay in astronomy with me, because of the internal pressures that I put on her. I tend to make quite strong demands about getting things accomplished, and it was just best that she do something else. We both understood that, and she's very happy doing something else.

Weart:

How do you think that the fact that you're a scientist has affected your marriage?

Sandage:

Well, not the fact that it's a scientist. The success or failure of a marriage depends upon the character and nature of the people. We both are quite strong in individualistic pursuits, and I think my being a scientist and her being interested in many things has been an asset.

Weart:

Has it had any effect on your children?

Sandage:

They see a father that works quite a bit. They have expressed great interest in the work. They both known how to computer program. They know how to measure spectra. They know how to observe at the telescope. They've expressed great interest in some sort of scientific intellectual career. I'm home every night that I'm not on the telescope; I maintain an office at home. So although I work at home, it's a very close-knit family, and I must say, a very happy family.

Weart:

I understand. I try to keep an office at home too. It's a little difficult when they're two years old.

Sandage:

Yes. (laughs) But we've been many places together. We spent 14 months in Australia.

Weart:

Oh, that's right, you went down to Mt. Stromlo.

Sandage:

We lived on the mountain, and that's been an experience for them. We go camping together. And they understand and accept the times that I'm away.

Weart:

OK, back to the social situation here, and particularly at Santa Barbara Street. Where and how in general do staff members exchange ideas about research? Here, we could talk about sort of from the fifties to the present, where they've exchanged ideas about research and how that's changed.

Sandage:

Well, there is a close-knit atmosphere at the observatory, and again, everyone by some means or other knows what everybody else is working on. That comes about by being together on the mountain, often; eating meals together on the mountain when you're scheduled. It also comes about because there are weekly colloquia on the campus at Cal Tech that most people go to. And then there are informal exchanges day by day. Most people tend to sit and work, but there's not a day goes by that you don't walk up and down the hall or drop in on somebody, and he will say, "Look, I've got something interesting." Or people will come in and ask what this machine is (plate measuring machine) and I'll show him what I'm doing. We have committee meetings of various types, to settle questions of priorities in the engineering department, what instruments are being built and time allocation committee meetings. There's probably a committee meeting every two weeks, and you then see the staff in this whole series of exchanges.

Weart:

I see. Has this changed at all during the time you've been here?

Sandage:

Santa Barbara Street has always been a very quiet place to work, and the amount of interaction that goes on varies from staff member to staff member. There have been times when the personal tensions were much greater than other times. For the last year or two, it's been a very happy place. The personal tensions have decayed; the divergence of views, for example, on the nature of the red shift, are accepted—not agreed to, but are not a cause of personal difficulties.

Weart:

Was this one of the main tensions in the past?

Sandage:

This, I think, was a great tension. It came about between Arp and myself, because of my rather conservative views of the proper style in science—the requirement of a lack of unsubstantiated speculation.

Weart:

I noticed that you did quite a lot of work with Arp in the fifties on clusters, and I wondered how that relationship changed.

Sandage:

We were very good friends. The difference in philosophy in doing science, at at time, was sufficient that the collaboration was not succeeding successfully, and it backed off, and backed off rather far, for a while. But the last two or three years, we accept each other's differences of opinions. Although he's very adamant on the non-cosmological nature of the red shifts, and I believe there's nothing to it, it doesn't matter what we believe, because nature is the way it is; if people are not convinced, it just means the data are not so overwhelming that everyone is convinced. And that means more work is required at the telescope.

Weart:

I see. And you feel that this divergence in attitude was clear even before the quasars, to some extent.

Sandage:

I think so, yes.

Weart:

Have there been other sources of tension here, or is this the main thing you're referring to?

Sandage:

I think that Hubble kept himself aloof from the staff. He was very highly respected, but did not interact with the staff. He and Van Maanen had a classical disagreement.

Weart:

I was thinking about the time when you were here, from your own experience and observation.

Sandage:

Baade and Hubble wre not active collaborators. They respected each other, but they worked more or less independently. Rudolph Minkowski was a great go-between, and a very jovial person, a very easy person, and caused the atmosphere to be very much lighter than it would have been if he hadn't been here. But I came in as a very young person.

Weart:

There was a whole crop of young people who came in in the fifties—in fact, the old people left.

Sandage:

That's right. There was a strong dichotomy, a generation gap of one whole generation.

Weart:

Did the atmosphere change with that generational changeover? or did the young people assimilate to the atmosphere?

Sandage:

The young people who came up here tended to be rather quiet also. Worked very hard, in a quiet way. The last three or four years, there have been a lot of young people come in, and that's changed the atmosphere of Santa Barbara Street incredibly.

Weart:

How so?

Sandage:

There are five Carnegie fellows and two Las Companas fellows now, and they're all under 30. They talk a great deal among themselves, generate ideas, help each other with the computer, and generally have livened up and lightened the atmosphere a lot.

Weart:

I see. I should interview one of those young fellows, just to get a picture of what things are like now.

Sandage:

Yes. Alan Dressler would be a good man. Free and open.

Weart:

I'll have to think about that. Has there been much after-hours socializing among the staff?

Sandage:

Not very much. Babcock is not like Harlow Shapley. And there is a more diverse group of friends on the outside, that are non-astronomers, that most staff members tend toward.

Weart:

I see. People from Cal Tech or just from the community?

Sandage:

Both.

Weart:

Has there been much discussion of subjects outside astronomy—biology, politics, philosophy?

Sandage:

Oh yes. You can't stop astronomers from talking about politics. There are conservatives and liberals here. It used to be that Santa Barbara Street was almost entirely conservative, and Seth Nicholson was the only Democrat amongst very many very conservative Republicans.

Weart:

Right, Hoover Republicans or Taft Republicans–

Sandage:

That's absolutely right. Humason was a strong Republican; he would assign the observing time on the mountain and he would make sure that every Election Day, he'd send a Democrat up the mountain so he wouldn't be down here and able to vote.

Weart:

But now you say it's more of a mixture?

Sandage:

I think it's more of a mixture, and certainly with the young people, it's more liberal.

Weart:

I see. I'm interested in other changes in the way things have been done here, and particularly in how the telescope assignment has evolved. I'm not sure when you first joined that committee.

Sandage:

That's a rotating assignment, a subcommittee of the observatory committee. The observatory committee is reconstituted every year.

Weart:

You were a member, starting in 1960 1 guess, of the observatory committee, so you've been on the time committee intermittently since then?

Sandage:

That's right. There are years that I'm not. I think I've been on probably two-thirds of the time.

Weart:

I see. Well, tell me how this committee functions?

Sandage:

It's been essentially the same procedure all the time. You have to justify the time that you are assigned. You submit to the director a proposal, with the amount of time that you request and the projects that are involved, and the proposals then go to the committee. The decisions are supposedly made on the basis of merit and past productivity. Now, merit means whether the science that is expected is good, and also in some sense whether it's current or relevant. That's very hard to come by—there are very few planetary astronomers on the staff, but clearly if somebody wants to do the occcultation of the rings of Uranus, that's quite relevant. So it's the way any project is assessed—whether it looks like it's good science or not.

Weart:

Is there any consideration given to whether it's particularly adapted for the instruments you have? A feeling, "This could be done as well at Kitt Peak," or whatever?

Sandage:

If we can do with our instruments, then there's no question about the competition on the outside. That is, if a staff member wants to do it, he presumably has thought out how he will do it, and if it's possible. The staff knows the instrumentation so well that no staff member would propose things that are impossible.

Weart:

I see. What about saying, "Well, you should do this on the 100-inch, not on the 200-inch," that sort of thing?

Sandage:

There is that looking at the proposals, and sometimes that is stated.

Weart:

Is it essentially the subcommittee, you argue it out on the subcommittee? or will particular people be responsible for particular areas?

Sandage:

No, all proposals are reviewed by everyone, and then there is a vote taken.

Weart:

A formal vote?

Sandage:

A formal vote.

Weart:

Has this always been the case?

Sandage:

Well, it's been more informal in the past when the pressure hasn't been so great. Also, guest investigators' proposals are put in with the staff's; everything is voted on. It's true that the guest investigator proposal has to be really very good. Since we're not a national facility, and we don't get money for guest investigators in that sense, we don't have the same responsibility as Kitt Peak. QUESTION BY WEART MISSING That's right. In fact, we're not compelled to give anything. We take the load off Kitt Peak and the other national facilities. It is an advantage to us, because it brings people in from the outside, and we certainly gain from that interaction. There's more time available on Mt. Wilson than there is on Palomar for guests; the staff demand is less.

Weart:

I see. So in the voting, there's a certain—?

Sandage:

Unstated bias.

Weart:

I understand.

Sandage:

But many guest investigator programs have been put on the 200-inch.

Weart:

Oh yes, I realize that. Is there more pressure on dark time?

Sandage:

Oh yes. And there's a great deal of pressure in the February, March, April, May time. That's when the North Galactic Pole, where the major parts of the galaxies are, is up.

Weart:

Is there a general feeling that the telescopes should be doing a lot of cosmology?

Sandage:

Well, there's a general feeling it's doing too much cosmology now, of a particular kind.

Weart:

This is a recent feeling?

Sandage:

That's right, in the last two or three years. There are two competing groups now. There's competition within the observatory. It took a little while to get used to that fact, but there are two competing groups doing the same problem, and that's the extension of the Hubble diagram.

Weart:

I noticed that, yes. I was going to ask about that. how did that come about?

Sandage:

I don't really know. The problem is of such interest that if there is freedom, anyone presumably is free to do anything they please.

Weart:

So in a way it's surprising that it didn't happen sooner.

Sandage:

Well, one might say that there was more interaction on a very personal level in the early days than there is amongst the young people now.

Weart:

There might have been more of a tendency to avoid what someone else was doing?

Sandage:

That's right, out of a feeling of that's the right thing to do. I think the attitudes toward doing science in a particular style toward others has changed, along with the general attitudes of society. I think that the building of fences around certain problems, which was certainly the case, and a more gentlemanly attitude of maintaining those fences, that's decayed in the same way that many similar things on the outside have. The world of astronomy is less civil than it was. The outside has become very much more familiar. And so the young people I think have taken it as a matter of course, and right, that there should be no restrictions on problems that they want to do.

Weart:

I see. This makes it difficult for the committee. I suppose one absents oneself one's own thing is being voted on.

Sandage:

Oh, that's right.

Weart:

Getting back to more to the generalities; if the vote is close, is there any attempt to make a consensus, or is it, "OK, the vote's over?"

Sandage:

Well, what happens is, you make the sums and you know how many nights are available, and those sums are known to the people requesting. The requests themselves are tailored so they're not outlandish. So the vote is, how many nights to be assigned to each person. Then those votes are tabulated. They're summed and compared with the total available. Then there's a factor applied across the board to everyone. If there is a factor of two oversubscribed, and they want to give 20 percent of the time to the guests, then we allocate 80 percent of the time to staff, and multiply every summed average vote by whatever factor it is to make it come into line.

Weart:

I see, nobody will be given 1.2 nights. Probably nobody will be given one night or two nights, it has to be above a certain level.

Sandage:

That's right. In fact, it's worked out that the demands and the available time are such that no staff member gets less than twelve nights a year on the 200-inch. But now there are many auxiliary instruments, the 100-inch, the 60-inch, and now the 100inch in Chile—we are really very fortunate at this observatory in having so much (mirror) area.

Weart:

The one committee allocates time for all the telescopes?

Sandage:

That's right.

Weart:

Now that you have this extra telescope it takes that much longer, I guess.

Sandage:

That's right.

Weart:

There is a separate set of proposals for the Chile telescope?

Sandage:

That's right.

Weart:

Probably less pressure on that because it's farther away?

Sandage:

At the moment, but it's only been in formal operation for a year. And since it's so remarkable and so unique and so powerful—it's more powerful than the 200-inch in the sense that the seeing is so superior, almost all th4 time, to Palomar—much of the Palomar requests will go to 100-inch at Las Campanas. The Palomar time should (then) be very much easier to get.

Weart:

OK. I don't quite understand when you say you vote to give a certain number of nights—do different people vote for different numbers of nights and then it's averaged—so it's just an arithmetical average?

Sandage:

Yes.

Weart:

I see. I'm going to get into thing because I think it's very important for modern astronomy, exactly what happens in that committee and how the time gets allocated.

Sandage:

It's amazing, the consensus that is present in these meetings, because everyone knowns, or has a feeling for, what he believes the worth of a given project is, in terms of percentage of total time. And that feeling of worth is surprisingly the same with everyone.

Weart:

So there's a fair consistency. Do you know how the other people vote, is it open?

Sandage:

No. No, you don't. It's a two-day session. We discuss every proposal. And out of that discussion, at the end of each discussion, then a vote comes, a secret vote.

Weart:

I see, you write it down?

Sandage:

On a sheet, and those sheets are given to the chairman, and he averages them that night. Then he comes in the next morning with the average number of nights that the committee has voted for this particular program, in order, number one, two, three, four, on the 200-inch, 1.00~inchl 60-inch, 100-inch in Chile ...

Weart:

I see. Are these things kept?

Sandage:

Yes.

Weart:

I see, so there's a full record. That's very good. Anonymous but that's all right. That would be very interesting for some future sociologist.

Sandage:

They're kept in the director's file; each of the committee members probably throws their stuff away.

Weart:

Well, so long as there's one full set. I can see some future sociologist will be very interested in this.

Sandage:

Yes. Kitt Peak is a little different and a little more interesting; each proposal is sent out by mail to a member of the committee, and before any discussion takes place, that member votes on a scale of one to five, and telephones in the grade. Then we all meet as a committee, having the votes tabulated and the proposals ranked.

Weart:

I see, so there's less opportunity to discuss them in a way?

Sandage:

We discuss those that have a big dispersion in the vote. There is a line drawn; everything above that can be assigned, and we look at those below. Then anyone can bring up a proposal, and those with big dispersions in the vote are automatically brought up.

Weart:

I see.

Sandage:

The line always goes down, and some of the people on top of the line get scratched out, because some things that the visitors propose cannot be done.

Weart:

So that's one thing that differs from here. How in general does the Kitt Peak committee operate, aside from these formalities, in terms of philosophy or approach? How do they differ from the committee here?

Sandage:

Well, it's quite amazing, the smallness of the blocks of time. People get five nights a year to do major project.

Weart:

I've heard complaints about that.

Sandage:

You just cannot do long-range programs at all. The only observatories that can do long-range programs, take a large amount of data, at the moment are here and Lick. The staff at CTIO^[20], where each staff member can be guarenteed maybe 30 nights a year, can also do it. But the staff at Kitt Peak and CTIO are burdened with building instruments and acting as service personnel to the guests. So the long-range, Hubble-like, Baade-like programs of Cepheids, where you have to accumulate data for ten or fifteen years, just can't be done. They can't be done for a second reason, that the committee looks upon the results of the last year at Kitt Peak before they assign time for the next year. So if it takes ten years to get Cepheid light curves, you have to have faith in the person to say, "You don't need to have results from last year."

Weart:

I see. You mentioned that here too they take into account not only the scientific merit but past performance. That might be difficult, though, with the younger people.

Sandage:

Yes, there certainly is a built-in bias. The Fellows are given all the benefit of the doubt, and they are treated very leniently. Let me say something now that is very troublesome to this observatory. We have very great financial problems because we're privately funded, and we do take the burden and act as a national facility in the guest investigator programs. We could do more, but the fact that the Mt. Wilson telescopes are mostly guests, and 20 percent of the 200-inch time and about 60 percent of the 48-inch Schmidt time is given to guests—not sold to guests, but given—is an anachronism, because we don't get government support to do that. And our technical backup is absolutely minimal.

Weart:

Right. And then of course with the new Chile telescope, it's that much more demand on available funds and backup.

Sandage:

That's right.

Weart:

I wonder if you could give me some examples that seem to you particularly revealing—usually when there's some conflict or excitement, that shows what people's underlying feelings are—and without mentioning names, things where there was a particular division of views revealed on these time allocation committees, either here or at Kitt Peak.

Sandage:

We don't know, at Kitt Peak, the later input and complaints of the proposers. There's always a problem of bias of the committee members, in programs where there may be personal competition for example. Now, certainly the really good scientists are often proposing the same things, and the very good scientists are on Kitt Peak allocation committee, so there's a possibility that that happens. That it does not,happen very often is reflected, because there are eight committee members, all with very diverse views, at Kitt Peak, and the votes are made before any discussion, and it's just incredible the degree of consensus before any discussion is made. We see the small dispersion. It used to be that there were eight columns.

Weart:

I see, so you could see what was going on.

Sandage:

You could see what was going on.

Weart:

They've saved these records at Kitt Peak, by the way, from the beginning. I checked on that. That's another interesting thing for the future sociologist.

Sandage:

The largeness of the committee acts as a brake on any personalities conflicts or bias in terms of the conflict of programs. Now, here it's much more internal.

Weart:

You know each other much better.

Sandage:

We know each other much better. The passions are very much higher. The views that "So and so is not doing good science, is an embarrassment to the observatory," have been expressed privately for years. But it's surprising that those people that you would think cause the difficulty are those that come out the best. It's like bending over backwards to make sure that on the outside there's an appearance of fairness.

Weart:

I see. "We won't deny him his nights, even though we don't think he's worth it."

Sandage:

That's what I'm saying now. But having said that, I do remember that there is unhappiness at the small number of nights given certain people, and I guess therefore what I said is not absolutely true. There was a period when these questions were not resolved fully, and there was dissatisfaction of several staff members.

Weart:

There would still be a feeling that there was an irreducible number of nights that you had to give somebody?

Sandage:

There is no guarentee of any telescope time at all in anybody's appointment. But there's an underwritten belief that if he's a staff member, there is an irreducible number of nights.

Weart:

I see. And of course because all the proposals are lumped together in one thing, you can't say, "Well, we'll give you nights to do this, but not to do that foolish program." You just give them the nights, or not.

Sandage:

That's right. And then the man can decide what he'll observe. There's great freedom about what is observed.

Weart:

He can change his mind and observe something different than what's on the proposal?

Sandage:

And he can observe what other people are observing. He can compete with other staff members. That's where a lot of the friction has started. But in an exciting field, who is to say whether that's right or wrong? Internal checks is the way science progresses.

Weart:

You have to do it. We'll get to that I think, when we talk about the quasars. Going up now to the Observatory Committee, and other decisions in general. I'm curious about the role of the director and the Observatory Committee. How are decisions made?

Sandage:

Well, very few things except matters of the most general policy are brought to the Observatory Committee.

Weart:

Is it like a board of directors and a company president?

Sandage:

It's a convenience of the director's, that he can use in any way he pleases, as a backup for decisions, as a protection for him. There are certain matters of interest that are brought up before the Committee, such as the possibility of making an airlinefree zone around Palomar. And if that's desirable, what moves of the FAA (Federal Aviation Authority) are required. The problems of living on the same mountain with together with the Forest Service. They want to open the trails, and what would that do to the observatory? Appointments are passed by the Observatory Committee, commendations for staff appointments. That's probably the most important aspect. But the homework has been done very thoroughly before it comes that high, and consultants have been made privately, so by the time things get there, it's almost an accomplished fact.

Weart:

I see, so the Committee would normally not disagree with the director.

Sandage:

The director, if he's a clever enough politician, will always make sure that doesn't happen.

Weart:

It's not always possible.

Sandage:

It's not true that the Committee doesn't ever disagree with the director. Some of the Committee meetings are quite heated.

Weart:

I see. Nevertheless the director has the final say?

Sandage:

The Committee, by the constitution, is only advisory to the director.

Weart:

I see, so even if everybody is adamantly opposed—

Sandage:

He can do what he pleases. If he thinks he can get away with it.

Weart:

I understand.

Sandage:

There have been staff revolts before. But he's gotten away with it always; he's still the director. There is a strong sense among some people that right or wrong, he is the director.

Weart:

Captain of the ship or whatever.

Sandage:

That's right, and only under the most dire circumstances do you put him out to sea.

Weart:

In fact, you couldn't, except by going to—

Sandage:

–the board of trustees, that's right.

Weart:

That's not a thing you would—

Sandage:

That's a major crisis, and I think most senior people realize that that has the seeds of their own destruction.

Weart:

Has that every been done by more than one or two individuals?

Sandage:

Yes. It's been done by the whole staff, occasionally.

Weart:

Do you want to tell me about that?

Sandage:

No, don't. (laughter) No. But it's come out right. it turned out that the director was dead right.

Weart:

OK. I can make a guess to what you're referring to. I'm interested in the institution of meetings of the whole staff. How did that come about?

Sandage:

There is in the bylaws now of the agreement between the two institutions^[21] a statement that the staff shall meet once a month, whether they have anything to discuss as a staff or not. I think we are probably the most independent research organization I know, in the sense that every scientist can do what he pleases. There are no staff programs and no observatory programs. And the budgets are so fixed that there's no freedom or airy spaces to look. And so the staff meetings normally don't have anything to talk about, except grave feelings of discontent of one type or another—if the mountains are being run right, or if personnel is right or wrong. But that normally is more effectively done in smaller groups.

Weart:

So the staff meetings are blowing off steam?

Sandage:

The staff meetings generally have become a pro forma meeting. The director can use them as means of general announcements to the staff that aren't done any other way. But we're so close-knit a group that almost everything that happens is known anyway.

Weart:

Known immediately, I see.

Sandage:

Who's appointed, to what, who is up for appointment, what the budgetary situation is. But it's now developed into a much freer and easier form; now the staff meets once a month for lunch as a staff, and you talk about anything you please. It's proved to be a very good way to get the rules satisfied.

Weart:

I see. What were the staff meetings like when it began? Can you tell me about the first meeting of the whole staff?

Sandage:

I think it began only four or five years ago. I don't think under Bowen

Weart:

-1970, 1971, something like that.

Sandage:

That's right. That was when there was a feeling that there was not to begin communication. At that time, (Las) Campanas was being built, and it was a forum which Babcock used to communicate the progress. That was a miraculous development, Campanas itself.

Weart:

Yes, I want to ask about that. Maybe you could talk about that now. You must have been closely involved in the sense that you were one of the senior staff members by that time.

Sandage:

Campanas, or a Southern site, was a dream for a long time. It was a dream of Hale in 1906.

Weart:

Yes. I shouldn't ask you to tell me the whole history, because that's already written down. What I should ask is about your involvement—what sort of involvement did you have with that effort, the fund-raising and grant applications and so forth?

Sandage:

It was in stages. I was very heavily involved, at the very beginning, in the concept, in the sense that when Bowen was director, Babcock and I went to Bowen and asked if he would object if we could explore the possibility of a Southern observatory station.

Weart:

How did that start? You and Babcock were talking about it?

Sandage:

Yes, and it was the feeling that this was the next move that Carnegie Institution should make in astronomy, that we had to do something as an institution, and the Southern Hemisphere at that time had no plans for telescopes. Bowen was reluctant, because Palomar had just gone into operation, and he felt that the whole energy of the staff and himself should be to make that work scientifically. But finally after some persuasion he said, "OK, go ahead." So a site-testing expedition was organized, entirely by Babcock, for about five years, many A mountains were explored, including Cerra Maralo. After a great deal of negotiation, which I don't understand, the agreement of AURA^[22] was not consummated.

Weart:

You were not closely involved in all that?

Sandage:

At the time when finally the project had gone so far as to have received the approval of the board of trustees of Carnegie, so that land could be bought, I was in Australia. I was not connected with the site survey. I did go back to Washington on two occasions and talked to the executive committee of the board of trustees about the need for a Southern Station, and why the astronomers thought this was important. Babcock and I both went Associated Universities for Research in Astronomy back on that. Babcock and ly when it looked like it was not going to happen, called up Crawford Greenewalt.

Weart:

This was after the Ford Foundation thing fell through?

Sandage:

Yes, that's right. The whole Ford Foundation thing Oh, such a painful experience that I've forgotten completely about that. The plan was to have a 200-inch in the South, and a great deal of groundwork had been laid with a man by the name of Borgman, who was the Ford Foundation field officer. Borgman was very sympathetic toward the plan. In fact, before McGeorge Bundy came in as president, the former president, and I don't know his name——Stratton?—and the outgoing president of Ford Foundation had agreed to the thing, and there was to be a transfer of stock actually made to the Carnegie Institution. Arrangements were being made between the lawyers to transfer the stock. Then McGeorge Bundy came in, and saw this enormous item, and through a series of moves, which I won't describe but which is understood here by a number of people, the Ford Foundation withdrew the offer, and gave the offer—together with the NSF funding, because Stratton was also the chairman of the relevant committee with the National Science Foundation—gave the money to build two telescopes, one in the North, one in the South but to AURA. After that blow, which really was a terrible blow to the project, Babcock stood fast, maintained the dream, and attempted to raise funding in another way. It was at this time, when everything looked impossible, that we called up Greenewalt to see if Horace and I could come back to Delaware and talk to him. Well, it happened that he was in Washington State, and he flew down to LA International Airport, and we got a private room and talked with him for about three hours between planes. He was very interested in these events.

Weart:

He was familiar beforehand with the various plans?

Sandage:

I expect he must have been, because the executive committee of the Carnegie Institution must have been appraised of the whole Ford Foundation problem. He then went back and talked with the family.

Weart:

In this meeting, you already hoped that he might be able to come up with some solution?

Sandage:

That's right. He was chairman of the astronomy subcommittee of the Carnegie Institution, and he was a trustee of the Carnegie Institution. We just laid out again the very strong hopes and dreams for a Southern observatory.

Weart:

Did you discuss the scientific merits once again?

Sandage:

Oh, sure. Being a scientist and engineer, he was very quick on the uptake.

Weart:

He could understand the value of the Southern skies.

Sandage:

Yes, and he certainly understood the excitement of a wilderness development in a good site and all that meant, and the challenge of that, and that it was the right emotional and spiritual—and interesting—thing to do. So out of that came the gift of the Greenewalts to purchase the land if a site could be found. That site was finally found by the site survey team. Horace (Babcock) had never given up, and the site survey work continued, and the land was then bought in 1969. It's proved to be an extraordinarily good site, just fantastic. It's a ridge that juts out into the prevailing wind, and there's nothing for 100 kilometers to windward.

Weart:

Perfect laminar flow.

Sandage:

That's right, even though it's over land. And it's dark; you don't see anything. The seeing, from experience, is better on the average than it is on the best times at Palomar, and it's good seeing most of the time.

Weart:

Half a second of arc or something.

Sandage:

That's right, it's just remarkable.

Weart:

You've been down there several times.

Sandage:

That's right.

Weart:

I can tell you love it.

Sandage:

It's amazing. That money was given for the site development. Then Henrietta Swope gave the bulk of her inheritance anonymously.

Weart:

Oh, I didn't realize that was the bulk of her inheritance.

Sandage:

That's what the belief is, anyway. Out of that came the 40-inch telescope. And when Greenewalt realized that then it was a viable thing, he was doubly generous and the family put in money for the Du Pont Telescope.

Weart:

Were you in on that again? Did you talk with him about the merits?

Sandage:

Yes. After the purchase of the land, I was not involved with the project very heavily until the Swope 40 inch telescope was built and on-site. I was heavily involved with the first years of that telescope. It was Horace Babcock’s tenacity, in the face of tremendous apathy of the staff. It is true that the challenges of the science with the 200-inch were great, and it was, an instrument already there, and if you didn't do the science in it then, why, the staff thought they would lose the opportunity. Horace put all his energies into Campanas, and the apathy of the staff continued, in general. There were a few people, like Arthur Vaughn on the optics, and then the whole engineering group was built up around Companas,with Bruce Rule and three of his engineers, and those people really carried the project for the 100-inch. And it's paid off.

Weart:

I see. Because they got an engineering enthusiasm.

Sandage:

That's right, and Arthur was interested in the optics. It is Horace's monument. The future will show that it was the right thing to do, that the staff was wrong in not enthusiastically supporting it. If Babcock were not strong and single minded and tenacious, it would never have come about. But that's also one of the reasons that the staff was apathetic; Horace tends to be very close with all that goes on, and so it's very hard for him to take the staff into his confidence. It was not like the building of the 200-inch, when everyone on the Mt. Wilson staff knew what was going on and were a strong part of it. I'd say a third of the staff were very strongly a part of it, and that includes the electronics people also.

Weart:

I see, and for the rest it was simply a diversion of—

Sandage:

–effort, time, money; and an apparent neglect of the other two mountains.

Weart:

Right. It's true, of course, that one couldn't improve the spectrographs or whatever on Palomar and so on, because the money was going to Las Campanas.

Sandage:

But in fact, that's a wrong statement. Because there can be no money exchanged between the two institutions—Cal Tech owns Palomar, and Carnegie money could not be put into Palomar. It could have been put into Mt. Wilson.

Weart:

It could have been put into Mt. Wilson, which of course is very old by now.

Sandage:

But Mt. Wilson is a crucial site. Mt. Wilson is a better site seeing-wise than Palomar.

Weart:

Except for the light, of course.

Sandage:

That's right, but there are many programs for which the Iight is not important.

Weart:

Well, another question is about what your relations with Cal Tech people have been like? How much do you see Cal Tech people, and when?

Sandage:

Well, there's a great deal of interaction among some staff members. I think that there's as much interaction as you desire, and it runs the whole spectrum, from not seeing anybody——some staff members up here don't see many people from Cal Tech. It's not a day-to-day elbow rubbing.

Weart:

What about yourself?

Sandage:

I tend recently to have been quite isolated from most of the group activities.

Weart:

There have times when you would go down there, and times when you haven't?

Sandage:

That's right. In the early days I used to attend all the colloquia. I used to teach on the campus; I taught three courses in the early days. I had students from the Institute.

Weart:

Had a few graduate students?

Sandage:

That's right, on pre-Ph.D. projects.

Weart:

This was only an occasional thing?

Sandage:

Only an occasional thing, that's right.

Weart:

Did it have much effect on your work, your teaching?

Sandage:

You always tend to learn an enormous amount when you teach. I learned a lot.

Weart:

What about these lunches that they have at the Athenaeum?

Sandage:

Friday lunch? That still goes on, and some people partake and some people don't.

Weart:

Have you attended that very much?

Sandage:

I used to, but I don't attend it very often. I guess, as I told you earlier, that the long term nature of what I'm going now is so well laid out in my mind, and it's going to take so much observational work that is of a routine nature, that in a sense, to be diverted continuously by the interactions and new ideas will jeopardize the reaching of those long term goals. I think it's true that there's so much activity in astronomy going on at Cal Tech that you could start Monday morning and spend all of your time until Friday afternoon in seminars and colloquia of an astronomical nature, and find out what other people are doing and the general nature of the field, and you'd never get anything done yourself.

Weart:

I see. It's the same thing as you were saying, that you don't read the literature that much.

Sandage:

I skim the literature. I used to think it was imperative to know everything that was going on.

Weart:

Did you read the whole ASTROPHYSICAL JOURNAL?

Sandage:

I used to read the essence. I mean, one can skim and go through a journal in twenty minutes and get the main points in the abstracts, the headings, and the diagrams and maybe one paper in each issue is sufficiently important for what you're doing that you have to owl study it in detail.

Weart:

But earlier you used to read more?

Sandage:

I used to readr in great detaily many different diverse fields, and I don’t do that much any more.

Weart:

I see. Would there be other journals that would be that important to you?

Sandage:

Well, the AJ; the MONTHLY NOTICES; I used to read NATURE; I used to read SCIENTIFIC AMERICAN; I used to try to keep up, as a matter of policy, on all those, but I found I just couldn't do it.

Weart:

I believe it.

Sandage:

And it's clear that my global view of science suffers because of it. The detailed view of what has to be accomplished in the next fifteen years will be diverted somewhat, clearly, it can't help but be. But I can't be diverted in a major way any more.

Weart:

I see. How do you usually learn about important new developments?

Sandage:

It's in the air.

Weart:

You learn it here, from people?

Sandage:

Yes, people come through. One talks about what happens at the colloquia. There's quite a lot of interaction, just day by day, and these things are just in the air.

Weart:

By telephone? Do people call you up or you call then?

Sandage:

Yes. I talk a lot with people like Spinrad at Berkeley, with Pierre Demarque at Yale, Ricardo Giacconi on the X-rays.

Weart:

I didn't kow you were interested. Just because of the cosmological interest of it?

Sandage:

Well, we've been personal friends for a long time.

Weart:

I see. So you just get on the phone sometimes and find out what's new. That's a change, I suppose, from the fifties.

Sandage:

No, I wouldn't say so. An awful lot of the underground knowledge is by the preprint circuit also. I get an enormous number of preprints.

Weart:

And you sort of glance through to see—

Sandage:

I see the titles, and the titles will tell you what the articles's about. You look at the diagrams and the tables and you know what the whole thing's about. You have to have a storage mechanism in your brainy and one calls it up when you need it for what you're doing, "Oh yes,there's an article about the absolute magnitude of RR Lyrae stars by so and so.''

Weart:

So you have to go to your preprint file and dig it out.

Sandage:

That's right.

Weart:

In terms of formal relations with Cal Tech, have you had much influence on decisions at Cal Tech that may be important to you?

Sandage:

I don't think so. The running of Palomar is a Cal Tech situation, and they do that all themselves. Through the observatory committee one has a leverage, but it's a very diffuse one. Those things that are really crucial are so common to all the staff, and thereys reasonableness about most of the staff, that nothing drastic ever happens.

Weart:

I see. I've questions about your outside relations. I noticed you were on the Whitford Report panel, I don't know if you've engaged in other things of that nature–?

Sandage:

Well, the Kitt Peak time allocation committee. Teri years ago I was a member of the Committee on Science and Public Policy of the National Academy.

Weart:

Yes, COSPUP. That was related with the Whitford Report?

Sandage:

No, that's the parent committee of all kinds of reports. I'm on the permanent organizing committee of the Solvay Conference, in physics. But I'm very inactive in all of these..

Weart:

You're not too interested in them.

Sandage:

No, I have withdrawn more and more in the later years, as the pressures to get things completed mount.

Weart:

You find the scientific work much more interesting?

Sandage:

Yes, that's true. And I realize that by asking myself, "Where would I rather be than where I am now?" anyplace I'm in. And the answer always was: I'd rather be back doing work in my office. So that was a clear statement that all the other things were done out of some other source of internal energy, than what I'm supposed to be doing.

Weart:

When you're observing on the mountain, would you rather be back in your office?

Sandage:

Yes, reducing the material I got the last run. Normally, the runs on the mountain are collecting nuts for winter; I'm reducing the spring and summer harvest ....

Weart:

I see, data reduction is the real harvest. In terms of the Whitford Panel, I was particularly interested in that. I wondered how that worked, do you recally preparing the Whitford Report?

Sandage:

Yes. I recall certain sessions that were really very heated and very interesting, amongst the radio astronomers. There was even at the time a strong feeling that if national centers were to be sponsored by NSF, the amount of money left over for private institutions and universities would be minimal. So there was a strong feeling amongst university people that the national centers should either be abolished, or not supported as thg place where all astronomy is to be done. And this was so strong amongst the radio astronomers that there was a revolt amongst them; the Whitford Committee was considered to be there forum, so that these views could be made more public. There was a meeting in Washington called specifically to address this question, when the complaints came in from the university radio astronomers. The director and associate director of NRAO^[23] were there, and it was a very heated two-day discussion.

Weart:

Between whom? The radio astronomers took this view, who took the other view?

Sandage:

All of the people connected with NRAO took that, the government scientists took that.

Weart:

I see, so it's not all radio astronomers, but it's among the radio astronomers, so to speak.

Sandage:

That's right. It was a debate on how radio astronomy as a science should be organized within the United States.

Weart:

The optical astronomers would have the same problem, when they're from private institutions.

Sandage:

That's right. But it was slightly different in that already two major observatories existed, Lick and Mt. WilsonPalomar, three perhapsy before Kitt Peak started. Kitt Peak was very small, and the optical astronomers were only from those two institutions. They felt more secure, in the sense that they knew where their funding was coming fromy and were not themselves depending on the NSF as strongly as the small radio astronomy departments at the universities were.

Weart:

I see. Of course, the final report did not list priorities, but were there problems about whether one should make a list of priorities, should one have more optical rather than more radio telescope?

Sandage:

Yes, and there were several meetings addressed to that. It was not as representative a study as the Greenstein one.

Weart:

Yes, it was a smaller group.

Sandage:

It was a smaller group, and more executive decisions were taken. In the Greenstein Panel there were five or six subcommittees. They made hearings around the country. We had a few of those, but our budget was very smally and most of the inputwas done by letter, and re-addressing the problems again by return letter.

Weart:

Were those letters saved? The Academy probably would have them.

Sandage:

I don't know. The man who was executive director of that, involved in the Academy, was Ned Dyer, I think.

Weart:

I can check that. OK, so although the net result was sort of a shopping list of good projects, there was some debate about what should be on the list, and whether some should be ranked above others?

Sandage:

That's right. And the question of small telescopes, like the 30-inch telescopes at all Midwestern universities, against a major telescope—that is, the concept of AURA versus individual departments—was a strong one. I can't remember at what stage Kitt Peak was at the time. I just can't remember whether it was in existence.

Weart:

It was certainly on the way. What view did you take—for a large telescope?

Sandage:

I would have taken the view that it does no good to put big telescopes in bad sites. And that was Irwin's view, against Edmondson for an example—at a Midwestern university. I don't know, I can't stack up all the other small Midwestern departments.

Weart:

The interesting thing is what views were held.

Sandage:

Herbig had a very strong view of elitism: that if You want good science, you put powerful instruments in the hands of the best scientists. If you want equalitarianism, where everyone gets a cut of the pie and you take your chancesy then you let everyone propose, you don't go to the good scientists and say, "Look, we need science done and you're the people to do it."

Weart:

OK. Well, what else in general about the national science? I don't know whether you've had any interaction with the space program or the national observatories. I'm talking now about the fifties and sixties.

Sandage:

Twenty years ago I was very heavily involved in committee work and the organization of science, and would go back to Washington very often. I was on NSF panels; I was on Academy panels. It would seem in retrospect that I was spending a very large amount of time doing that sort of thing. That was the time of disenchantment, whether I'm going to really do what I want to do, which is the research problems; I couldn't do both.

Weart:

I see, so you made a decision.

Sandage:

It was a very deliberate decision, to stop doing that sort of scientific politics.

Weart:

I see. Were you every on any Defense Department panels?

Sandage:

I was on a panel about the shooting of a rocket to disperse needles.

Weart:

Oh, West Ford.

Sandage:

West Ford.

Weart:

The Townes Committee? Or was that later?

Sandage:

Shapiro was in there. Bruno Rossi came to testify. it turned out that this was to act as an antenna for Defense Department signals to push the bomb button in Europe when other communications failed. We didn't know that at the time.

Weart:

This was before they'd sent the rocket up?

Sandage:

That's right. This was a committee empaneled to see the risks involved. There was a great deal of debate and calculation of the lifetimes, the decay properties of the orbit, the surface brightness of the belt. There were five or six meetings on that, and there was a big report written, when I was on that panel.

Weart:

Which concluded I guess that it was safe, that one experiment, but not necessarily the whole system.

Sandage:

There was a minority report also.

Weart:

Which side were you on?

Sandage:

I was on the minority.

Weart:

I wasn't aware of that. The minority report wasn't in the press.

Sandage:

Then the question of detection came after it was shot up, and we mounted an experiment, an observation, to detect it.

Weart:

Optically.

Sandage:

Optically, and we did detect it at Palomar, with the 20-inch telescope. It turns out that we were the only ones that detected it. We detected the rocket casing, and then the coma behind the rocket casing. It was absolute luck in some sense. we, had the orbit calculated, transmitted, so we knew where to point, and we kept the telescope on that pointy and it went through the field of view of the photometer. We published an article in SCIENCE about thaty Charles Kowal and myself.^[24] We reported the scientific aspects of the thingy and in the last paragraph went into the politics on the experiment, which was again the minority report. It was not right to have included politics, I thinky in a scientific report. There was a rejoinder from Shapiro, who wrote and said he was very pleased that we had detected this, but, "I'm very disturbed at your last paragraph."

Weart:

You weren't involved in any of the subsequent committees on West Ford?

Sandage:

No.

Weart:

Did you ever do any military research?

Sandage:

Never.

Weart:

Ever done any consulting for industry?

Sandage:

No. When McDonnell-Douglas was asking for the bid for the Space Telescope, I was approached by McDonnell-Douglas to be a consultant to them. But they did not get the contract, and that fell through. (Brief pause)

Weart:

The next thing is the quasar story and so forth, but first, because it's related, I'm interested in the growth of radio astronomy at Cal Tech. Did you have any connections with this?

Sandage:

Not with the growth. There were certainly connections between the radio astronomers and the optical astronomers. When Bolton was here, developing Owens Valley and the Cal Tech radio facility, he was strongly interactive with the optical astronomers. The quasar story really begins from that interaction. Tom Matthews, who was a staff member of the Owens Valley Observatory-he's not been given much due credit for the discovery of the quasars, but it really was very crucially dependent on what he had begun and what came of that.

Weart:

Oh yes. I guess I should asky even before you start on that, what can you tell me about Minkowskils identification of 3C 295?

Sandage:

That occurred as part of his general identification program for 3C sources. At that time, the positions of the 3C sources were really very poor, so the percentage of success was bad. But it happened that 3C 295 did have a pretty accurate position finally. I forget the circumstances on how that position came, whether it was from the 3C revised source list or whether he had an updated list somehow.

Weart:

Didn't he get information from a couple of groups?

Sandage:

He was receiving information all the time, from the Australians and from the Cambridge people.

Weart:

I see, was this because he had access to Palomar essentially?

Sandage:

That's right, and he and Baade acted kind of as the optical liason between those groups, and both groups trusted Minkowski. He was scrupulous in not divulging any information of one group to the other. There was very intense rivalry at the time. So he was privy to information on precise positions, and I just don't know where he got the position for 3C 295, whether it was special or not. But he asked that photographs be taken of the field. And I took those photographs with the 200-inch.

Weart:

Is that so? How did that work?

Sandage:

There were often requests, when people had a lot of time, for staff members to take plates for other staff members. You asked earlier whether there was very much taking of plates for other people, and I said, not very much. But there was some.

Weart:

This would be because this was a short time scale thing, had just gotten the position.

Sandage:

That's right—

Weart:

And you were the next person with a run?

Sandage:

–and it was perhaps disappearing into the west, and he didn’t have another run. That was fairly often done: one traded plates.

Weart:

I see; it would only take an hour of your time or something like that.

Sandage:

That's right. I don't remember the chronology exactly, in the following sense. There were two sets of plates that were required. There was a set of plates required first of the radio position, with the Ross lens in at the prime focus. This gave a big field. Then, after the identification was made, since the Ross lens distorts, the spectroscopists, before they went up for a prime focus spectroscopic run, needed blind offsets, from a bright star, because you could not see the object. So another plate, which was called a trail plate, is needed, without the Ross lens; you photograph the field and then stop the drive of the telescope, and the stars move, showing you the east-west direction, and that means then that you have the orientation and then the delta X and delta Y.

Weart:

I see, a few of the brightest stars will leave trails.

Sandage:

That's right; from the long exposure, say 50 minutes, YOU see the object, and then from the trail of the bright starsy you have the precise orientation. And by measuring the trail plates in Pasadena, you can determine the delta right descension and delta declination, from a bright star. On the spectrograph there are precision screwsp on the guiding eye piece. So I got the trail plate for him, and I don't know whether I go the direct or not. We could look it up if it was important. It's in the logs. It was fairly early in the season, because he then had two runs on it. The first run, he got a spectrum of I think a four-hour exposure; it showed a continuum, and it showed an emission line, which was marginal. It could have been a plate defect, but if it was real, it was 3727 and the red shift was large. So he had to get a confirmation plate.

Weart:

This was the same night?

Sandage:

No, this was the next run, as I remember it. I think he came back to Pasadena, and it was his last scheduled run on the mountain before his retirement. I think it must have been May or June of whatever year he retired.

Weart:

1960, I've got.

Sandage:

Or 1959. So he went up again; he had either three or four nights scheduled, and the first two nights were cloudy. The next two were clear, and I think he exposed a night and a half on the confirmation plate, and he stopped theny because the object was setting, about midnight, and developed the plate, and the confirmation was on that night. I happen to have been on the mountain at that time. The night assistant on the 200-inch was very cognizant of what was going on; his name was Robert Sears. We sat in the library while Minkowski developed the plate, and we could tell from the sound of his footsteps coming down the corridor that it was a success, because he was just bounding down the corridor, and he opened the library door with a bottle of bourbon in his hand and three glasses.

Weart:

You knew he was trying to see whether in fact, the line was there.

Sandage:

Oh, of course. It confirmed what held gotten the run before. The last observation he ever made with the 200-inch.

Weart:

It was quite an observation. Why would there be two astronomers on the mountain at the same time?

Sandage:

It happens, in general, if you're scheduled on another telescope, the 48-inch or the 20-inch photoelectric one. I must have been scheduled on one of those.

Weart:

By the way, I should ask you—mention of clouds reminds me—if a person has had bad luck with previous runs, does the scheduling committee take this into amount?

Sandage:

No, you just take your chances. It's an absolute loss.

Weart:

OK—quasars. Just to run down the chronology briefly, 1960 was the identification of 3C 48 with a starlike object. 1961, your paper, Matthews and SANDAGE-

Sandage:

–no, 1963.

Weart:

Oh, OK, sorry. I guess there wasn't anything published, but this was the time you observed 3C 196 and 3C 286. Maybe it’s in the Director's Report. You, and also Schmidt, took spectra. Then 1962 was when Schmidt took the 3C 273 spectra, and then the red shift was determined early in 1963. So that's the raw chronology, as I understnd it, but I'd like to go over what your role was in all this.

Sandage:

Well, Minkowski had started the systematic identification of radio sources in optical wavelengths. He was not very successful, because the positions were so poor up to that time. He and Baade had made this breakthrough in identifying Cygnus A and Virgo A, which is 4486, in two fundamental papers in the APJ about 1957. He continued the work of identification as the positions got better and better, but there clearly had to be a fundamental increase in accuracy. When Dolton started Owens Valley he understood this, and the two antennas which are acting as interferometers began to produce very precise positions. The Jodrell Bank people had also started interferometry, and were getting very precise positions. Tom Matthews, who was a staff astronomer at the Owens Valley Radio Observatory (operated by Cal Tech) began to assemble a list of objects that had very interesting radio properties. Jodrell Bank..—. had determined angular diameters for some of the sources and hacl begun to resolve the radio structure by their movable antenna, that went across the island of England, and Tom had access to these data. The principal man was the English radio astronomer (Henry) Palmer, and Palmer also has not been given very much credit in the quasar story either. It was his angular diameters and the lack of resolution A certain of the 3C sources, combined with the fluxes, that gave very large surface brightness to some of the radio astronomers.

Weart:

I didn't realize that, I hadn't known about Palmer.

Sandage:

It's not in any of the literature.

Weart:

How do you know about it, from discussions with Matthews?

Sandage:

Everyone knew about it at the time. At the first Texas Symposium on relativistic astrophysics, Palmer came over and discussed the technique and produced the list of small diameter sources.

Weart:

I see, so it was essentially this that got Mattews oriented towards these particular objects.

Sandage:

That's right. Matthews then compiled a list of objects whose radio surface brightnesses had only lower limits. They were brighter than a certain amount, and they were very high. Nowy this list was an extraordinary interesting list, because in retrospect, every quasar that was a 3C source was in Matthews' first list.

Weart:

I guess that's a sign of Palmer's diligence, to do all these angular diameters.

Sandage:

That's right. And it was Matthews' beginning of a program. Since the thrust was so strongly for optical identification, and Matthews was in the forefront of that, Matthews began to assemble not only this list of high surface brightness, but precise positions for them. At that point, Tom (Matthews) came to me and asked if I would be willing to join him in an identification program. I said, "Sure." So I took the plates, and he, with the precise radio positions, determined whether there was an optical identification or not.

Weart:

I see, this is actually fairly easy for you, because it's just slipping in a plate on your regular runs.

Sandage:

There was absolutely no science involved. There was no foresight-

Weart:

It takes a little time away from your other work.

Sandage:

That's right, but there was no conscious effort; all the ideas were generated elsewhere. And so, in August or September of 1960, 3C 48 was photographed, and I brought the plate back and gave it to Tom, and he said a few days later that there is, in the error box, a star. So in October I did photometry of that star, with the photoe ect c photometery and found that it was abnormally ultraviolet and further that it varied in optical brightness from night-to-night. Then I took the photometer off and took a spectrumy and it was very strange, unusual spectrum.

Weart:

At the prime focus?

Sandage:

With the Humason nebular spectrograph, the faint object spectrograph he used.

Weart:

And of course as soon as you developed it, you saw that it was a little weird.

Sandage:

Yes. It was like nothing that had been seen at that time. I think I took a total of some five or six spectra, and came back and measured the positions of the lines, and it made no sense at all. It was the furthest thing from anybody's mind that they could be red shifted lines, because before that time thre had been no object ever known with red shifts that was star-like in appearance.

Weart:

It was conceived as a radio star.

Sandage:

Star, that's right. I went to Bowen with the line lists and the spectra, after fiddling with the lines; Bowen was an expert on the high excitation forbidden lines at peculiar wavelengths, and he couldn't make sense out of it. He said, "Let's go down and show it to Greensteint who's been working on very high excitation lines of oxygen VI and so on in hot stars." So we did that, and Jesse couldn't make anything out of it. In the meantime, I'd been photometering the object and it continued to vary in light.

Weart:

How did you notice that, was it just a normal thing as you did more photometry?

Sandage:

More photometry.

Weart:

Why did you go back and do more photometry, once you had done the first?

Sandage:

It was a "star." Many stars vary.

Weart:

I see, so you were looking for variability.

Sandage:

Just to see whether in fact it did. The object was followed then for several months before it set. In the meantime, (3C) 286 and 196 were found in the same way. The optical positions were determined by Matthews on the basis of plates I had taken with the 200-inch. Then we measured the intensity in three or four other wavelengths and got a rough broad-band energy spectrum, which was a power law spectrum, and didn't really do anything with the material until 1963.

Weart:

Why was it that you didn't publish the spectrum?

Sandage:

Because I didn't understand it. And I'm so conservative that I won't publish anything unless I think I understand something about it. That's my downfall, in a certain sense. That's being so conservative that you don't believe anything is new under the sun. I tend to disbelieve almost every new discovery that's made.

Weart:

So you're on the conservative side of the inevitable conservative-liberal divide.

Sandage:

In science, that's clearly true, in politics not. Tom and I then had rough energy distributions and spectra for 3C 48; 1 don't think we had spectra for the other two, I can't really remember. Maarten (Schmidt) then began to get spectra of 286 and published a note, and then 147, 1 think, before he got the red shift of 273. Tom and I then were writing the paper up, as if these were identifications of three radio stars.

Weart:

That's right, you had started to write the paper up finallyr even though you hadn't understood it.

Sandage:

That's right.

Weart:

You finally decided you might as well.

Sandage:

That was three years after, in 1963, and during the process of writing up, Maarten made this fundamental discovery.

Weart:

Did you regard this as a very important problem at the time? Did you spend a lot of time trying to interpret this spectrum?

Sandage:

No. I considered it to be absolutely a side issue.

Weart:

So you didn't feel that Maarten was poaching on your territory?

Sandage:

Oh, of course not. I think at the time that the push for Cepheids in 2403 was really getting underway in a positive sense, and the extragalactic distance scale problem was beginning. And I thought that my getting plates for Tom Matthews was a diversion. Tom Matthews has not been given credit in the literature for this, nor has Palmer, and I think there's really a point to be made, that discoveries like this come only after the groundwork has been laid by a large number of peopley and history is very unfair.

Weart:

Yes. Looking at it as an historian, in preparing for this interview Matthews' role became very clear to me. I guess I hadn't pushed it back far enough to see Palmer's. That's very interesting, I'm glad you told me about it.

Sandage:

The early Palmer stuff can be seen in the publications of the first or second Texas Symposium on high energy astrophysics, also in MNRAS.

Weart:

Right, it's just that I hadn't taken enough time to go through it completely. About the variability of 3C 48, and the others, you mentioned how you got started on this. There wasn't any mention in your first paper on this argumenty which is very familiar nowy that the system could not be larger than the light travel time indicated.

Sandage:

We didn't make that argument.

Weart:

I wonder when you first heard this argument.

Sandage:

I think at the Texas Symposium.

Weart:

What was its impact?

Sandage:

In fact, the variability of 3C 48 was a long-term affair.

Weart:

Yes, although you had noticed some fairly short-term things.

Sandage:

I'm not sure we believed those. We said 15 minutes, I think, in the paper.

Weart:

Or was it only a possibility? Your thinkg showed some sort of week-long variations. That's still pretty short for something that's at tht red shift. That hadn't struck you until the Texas Symposium?

Sandage:

That put a restriction on the size.

Weart:

It's a familiar argument nowadays, and yet in fact I've found that it wasn't familiar.

Sandage:

I don't know who made the argument, in fact. It may have been Terrell, it may have been Burbidge, it may have been Morrison.

Weart:

I think Tommy Gold actually was one the of the first to use the argument, even before quasars, but it wasn't well-known certainly by anyone. Tell me a little about that first Texas Symposium, what the effect of that was.

Sandage:

I can't really remember. I don't remember whether the quasar story was reported there or not.

Weart:

Oh yest that was the first big symposium after the quasars. I guess it was the summer after the first report.

Sandage:

I can't remember.

Weart:

OK. I just wondered whaty in general, was the initial impact of the energy problem of the quasars.

Sandage:

I have never believed otherwise than that they were cosmological. First of all, when I learned that Schmidt had determined a red shift of 273, by telephone from Cal Tech, I first didn't believe it.

Weart:

He phoned you up?

Sandage:

There was some electricity in the air. I don't know how it all came about, but within a period of a day, there was real excitement every place. I took the plate of 273 upon which the identification was made.

Weart:

You took it for Maarten?

Sandage:

No. For Matthews! And that was the plate that Maarten and Jesse published unbeknownst to me. They did not communicate with me. Either Schmidt or Matthews made the identification from that plate, on the basis of the position determined both from the occultation in Australia and some other position they had. Now, I didn't believe that the identification was with red shifted lines, at firsty because it was so foreign. Everything I had ever understood about red shifts was connected with galaxies. When the evidence became unequivocal two days later, and then Matthews and Greenstein used my wavelengths of the lines plus the later measurements by Greenstein from the spectra of 3C 48 and found the coincidences of the lines withfItill greater red shift, and it all fitted, then of course one had to believe it. But then they had to be extragalactic, for me. There's never been a time that I could understand a mystical red shift with no physical basis. It was clear immediately that it could not be gravitational, by all sorts of the most elementary arguments.

Weart:

Within the first few days?

Sandage:

That's absolutely right. The story was very unclear for several years, as to what, in fact the parent objects were, whether they were isolated bundles of great energy—but the radio properties of quasars and radio were soon shown to be identical, in the sense of the separation of the double sources being interspersed. That is, you could not tell on the basis of radio the difference between quasars and radio galaxies. And by that time we knew enough about radio galaxies—we knew a large amount about radio galaxies—so that (it was clear that) there was some connection in the mechanism because they were double radio sources and so on. Then gradually, as the Hubble diagram of quasars emerged, and they were all brighter than the Hubble line, for radio galaxies–the radio galaxy line formed the lower envelope of the quasar distribution–that was a clue immediately that they were connected. Out of that came the belief that they were events in the nuclei of the ellipticals.

Weart:

Before we get into that specifically, I’m interested in your work with Lynds on the H alpha filaments in M82, and the suggestion that there’s a colossal explosion in M82^[25]. This was sort of happening at the same time that these quasar plates were being taken. I wonder what the relation might be?

Sandage:

That’s right. It’s very interesting to m, now Spencer, that people have made the connection historically of an explosion in M82 with a radio explosion in radio sources and double radio sources. That never entered my mind. M82 was a radio source, it was a very weak radio source, and it was not a double radio source, it was a core source.

Weart:

Where did the idea come from that this was an exploding galaxy?

Sandage:

From the kinematic field alone. The spectrograph was put along the minor axis by Lynds, and Lynds found that the lines were tilted, a 200 kilometer a second difference in velocity between the north halo five minutes out and the south halo five minutes out. And since the galaxy was almost on a line of sight, that meant and enormous–

Weart:

–radial expansion–

Sandage:

–on both sides.

Weart:

So there was simply no way out of it at all.

Sandage:

There seemed to be no way out of it at all.

Weart:

Was this the first time that it had seemed to you that it was possible for a galaxy to explode? What was the effect, to think that this kind of thing, this kind of explosion–?

Sandage:

Well, since 1918 M87 had been known to have a strong linear jet from the center. Curtis discovered that, if Michigan. And that had a bright knot plus four resolved knots in a straight line, and it was a natural phenomenon to believe something was connected with the nucleus, and shot out from the nucleus.

Weart:

Nevertheless, it was generally thought the early radio galaxies might be colliding galaxies or something like that. The idea of an explosion was still foreign to most people.

Sandage:

But the fundamental discovery was made many years earlier by Hanbury Brown and Das Gupta. They discovered by radio interferometry the double source nature of Cygnus A, and that was one of the fundamental discoveries in radio astronomy.

Weart:

I see, so it didn't seem like a tremendous–

Sandage:

–not at all. I was very pleased at the impact it apparently had, but it didn't seem like a very important discovery to me.

Weart:

I see. What about the growth of the idea that exploding galaxies are actually fairly common? Did this come before or after your work on quasi-stellar galaxies?

Sandage:

Well, "quasi-stellar galaxies" is an unfortunate term for radio-quiet quasars.

Weart:

That's right, OK, I'm using the original term.

Sandage:

Right. You really have read the literature. That's terrific.

Weart:

Well, I want to really go into this paper on "quasi-stellar galaxies."^[26] I haven't been asking about all your papers; I try to pick a few high points. The paper on radio-quiet quasars—

Sandage:

—did you like that paper?

Weart:

I liked that paper.

Sandage:

Did you really?

Weart:

I particularly liked it because ......

Sandage:

—it showed too much enthusiasm.

Weart:

No. I liked it because it was written in the order that the things happened.

Sandage:

A case history. (laughter)

Weart:

That's right. It describes how the discovery was made.

Sandage:

I became very conservative after that paper was published.

Weart:

I think it gives a good opportunity, if you don't mind going into detail of an example of how things are done. We can't do that for every paper you wrote, but let's go into this one. As I understand it, it all began from your search to identify quasi ...stellar sources. You were looking for objects with ultraviolet excess, and you found extra ones on the plates?

Sandage:

That's right. After the first few quasars were discovered, from Matthews' list, the radio positions still were very poor in general for the whole 3C (catalog). I decided to exploit the ultraviolet excess that was discovered in the first three by photometry—3C 48, 186, 296. 1 really forgot the technique that had been invented by Haro and Luyten; in fact, to made the same type of survey, using the 48-inch, by taking three-image plates. I took three-image plates near the positions of the 3C radio sources with the 100-inch on Mt. Wilson: ultraviolet, blue, and yellow. And in inspecting those, instead of finding one object with ultraviolet excess near the middle, on every plate I found four or five objects. At first I thought that my color balance was wrong. Then I checked a few of those photoelectrically, and sure enough, they had strong ultraviolet excess. I also forgot that white dwarfs had the same general characteristics. Daniel Harris had discovered that in 1950, from the UBV photometry of white dwarfs. This program was in conjunction with Ryle, who was now sending precise positions of other objects that his big long-baseline interferometer was producing..

Weart:

He was sending them to you'?

Sandage:

That's right. So Ryle and I published a short note on the optical identification of four more objects—3C 9 was amongst them—as a note in the APJ.^[27] But out of that program with Ryle came the finding that the surface density of ultraviolet objects is really very high. And :1 was puzzled by that.

Weart:

Had you been aware of the Haro, Luyten objects, or when did you become aware of these?

Sandage:

I remembered them quite a bit later, may by four or five months later.

Weart:

You just noted these objects and then let it go for a while?

Sandage:

That's right. I was very surprised at the very large number of objects that was "noise" in my sample.

Weart:

You figured that was a problem for the stellar people, so to speak.

Sandage:

No, I didn't make any hypothesis about it. Things I can’t understand immediately, I put in the back of my mind. But the connection was more interesting than that, because in 1952 there was a symposium called "Cooperation in Galactic Research," an IAU symposium held in Sweden, and this symposium was concerned only with stellar problems. There Haro talked about the survey that he and Luyten had just completed with the 48-inch, finding all these blue objects. There had been a lot of discussion of the blue objects. Zwicky and Humason had a list of maybe 4C) that he found with the 18-inch. Haro and Luyten had thousands.

Weart:

You heard about it at the symposium?

Sandage:

At the symposium. No one understood what they were. There were speculations that they were B stars in the halo, halfway because they went very faint; it was between us and Andromeda (M31), because they went very faint; it was believed that anything Mat blue had to be a B star at absolute magnitude -6, and so it had to be very distant. So Zwicky manufacturer a halo of our Galaxy that was enveloping the Andromeda nebula. He claimed that the halo of our galaxy was extending as far as M31.

Weart:

I see, and you heard this from talking directly with him?

Sandage:

It was common knowledge. He would give colloquia, he would write these in the annual reports of the director, and that was the claim—that the Humason-Zwicky stars were the brighter sample of the ones that Haro and Luyten were finding in great numbers fainter, and therefore the halo had to be very extended. At this Swedish symposium, the blue stars posed such a problem for what had been known in stellar distribution before, that that was stored in the back of my mind also. At that symposium I told myself, "I want some day to understand the blue stars.''

Weart:

That's so different from your previous interests, in a way.

Sandage:

Oh, I started out as a stellar astronomer, counting stars in Perseus. And the halo problems and the problems of galactic structure were very interesting. I didn't make the connection between the two sets until after the survey for the Ryle identification was well underway. Then it was clear that what I was finding was exactly what Haro and Luyten and Humason-Zwicky had found, many years before. They were blue objects. Then, at the second Texas Symposium, I reported on these new identification. And from the floor, a question was asked by Philip Morrison, "How do you know that these blue objects are not quasars in their own right?'' I said, "I don't."

Weart:

So it was back to the telescope?

Sandage:

It was back to the telescope immediately. And by golly, that was what they were.

Weart:

In your paper you give that in fair detail. You and Philippe Veron. Who was he?

Sandage:

He came over from France and worked with me for a year on optical identification problems of radio sources. When the work on Mt. Wilson with the 100-inch for the ultraviolet excess was going too slowly, I decided to take the search to Palomar. Not remembering the Haro-Luyten survey; I hadn't made that connection. There was a field that was photographed in three colors, to see how many blue objects there really were, and I (on the 48-inch) asked Philippe Veron to inspect it. He came back saying, "There are just tremendous numbers of blue objects.'' We should have understood at the time that that's just what white dwarfs would have looked like. But if I'd understood that, I would have told myself, "Well, they're all white dwarfs and you might as well not go ahead." I forgot about the white dwarfs. And with Morrison's comment, then, it's described in the paper that you referred to how the spectra were gotten. I told Maarten of the suspicions. I had gotten photoelectric photometry for a number of these interlopers in the meantime, and they were all very peculiar.

Weart:

Right, according to the paper, even before the spectra You did a (U-B, B-V) diagram.

Sandage:

That's right.

Weart:

This is just following up the photometry, so to speak.

Sandage:

That's right.

Weart:

From what you've said now, I know you already had in your mind that these might have been quasars. But still it's interesting that when you had this U-B, B-V diagram, you looked at it and instead of seeing a random scatter, you saw stars along the normal line, plus a fainter, different set—you noticed that you had two types of objects there. Was it at this point that you began to draw the connection with Haro-Luyten–?

Sandage:

There had been no photoelectric photometry of the Haro—Luyten sample. We had no idea what the diagram of the Haro-Luyten sample was. But the bright blue stars, the Humason-Zwicky stars, and then the Feige blue stars, were all very bright.

Weart:

The what blue stars?

Sandage:

Feige. Jacques Feige made a separate list of bright blue stars. And those were all bright, brighter than 14; they all fell on the normal Main Sequence. That's still a very fundamental dichotomy: the bright blue stars are Main Sequence stars, and the faint blue stars fall along the black body liner or above it.

Weart:

You noticed this'?

Sandage:

In my mind I then said, "All the Humason-Zwicky stars are nothing unusual. They're all bright, and the Zwicky argument of a vast halo doesn't hold, because it's a different kind of object.''

Weart:

You recognized that at this point.

Sandage:

That's right. The dichotomy, brighter than 14 and fainter than 14, was a fundamental part of what I thought was the argument.

Weart:

In a way it was stimulated by thinking about Zwicky's argument?

Sandage:

Yes, because that seemed so unlikely, as an explanation, that other explanations of the blue objects had to be correct. Now, it was not horizontal branch stars of globular clustersy because they would fall on the normal sequence. So it had to be completely different. And again, I forgot completely about the white dwarfs.

Weart:

OK. The next thing you say in the paper is, doing an integral county number versus apparent magnitude, you get a plot which has a change of slope. Why do that sort of a plot? Where did this idea come from?

Sandage:

Because the distribution of things extragalactic had been proven by Hubble to have a slope of .6 M nearby, and then when the red shift takes over, a less steep slope.

Weart:

I see. In other words, when you did that you had clearly in mind the idea of looking for very distant objects, extragalactic ones.

Sandage:

That plot came after the spectra were obtained. It was a justification of why I thought most of them were extra galactic. The hypothesis first was that they would have red shifted lines, like the radio quasars. And that was proved to be correct in a subsample by Schmidt and then later by me.

Weart:

In the paper you say that you had predicted that they would have high red shifts. When you took this spectra, you hoped to find high red shiftsy or asked Schmidt to take the spectra.

Sandage:

I guess I can't really reconstruct that now. That would still be true if the analogy with the radio quasars would hold. if the prediction was on the basis of the N(M) plot, then what I just said was wrong, that I had made that count to see whether it was feasible that they could be extragalactic. I can't reconstruct, in memory, which came first.

Weart:

OK. In the papery you described that you asked Schmidt to observe some candidates during his run. Then you had a run shortly after that which was somewhat foggy but nevertheless you got some spectra.

Sandage:

Yes, that’s a true statement.

Weart:

And found that they were in fact red shifted. It must have been exciting.

Sandage:

Yes. That was the most exciting month of my life, I'm sure.

Weart:

Tell me a little more about that.

Sandage:

Well, because if it were true, that meant that the quasar phenomenon was much more general. The surface density of these things was some thousand times that of the radio ones.

Weart:

So you saw that even before you had the red shifts.

Sandage:

Oh, of course. If in fact these were radio-quiet quasars, then the number of them—unfortunately in the paper I generalized and said, "All of the blue objects are extra galactic.''

Weart:

All the ones fainter than whatever.

Sandage:

Herbig's dictum is right—"If you find something exciting, wait a while, and it will disappear either in part or wholly."

Weart:

At least some of it will disappear.

Sandage:

That's right. Not all of this disappearedy in the sense that the surface density of radio-quiet quasars, if you go faint enough, is very high. And the white dwarf population does drop out. So all of the objects fainter than 18 or 19 are (quasars).

Weart:

Right. There's a few in between tbre that could be either way.

Sandage:

There's a number at 16 that are white dwarfs.

Weart:

You also mention in a footnote that there was a letter from Iriarte, that you got after the observing list for the Palomar run was made up, noting that Ton 730 looked sort of like a quasar. Do you have that letter still, by the way?

Sandage:

I probably do. [AS: Please do save things like this!]

Weart:

OK, so you have saved some letters, that's interesting. Did he write because he hoped you would observe ity that sort of thing?

Sandage:

I really forget. He may have been at the second Texas Symposium, when it was clear what was going to happen. I meany it was clear from my presentation of the finding of the large surface density of the interlopers, and in the possibility that these were radio-quiet quasars, that that observation was going to be made.

Weart:

I see, so in a way you were in a race with the world to——

Sandage:

I don't remember it that way. But I was certainly in a race with myselfy to see whether quasars were a small percentage or a fairly large percentage of the blue star population.

Weart:

I see. The run was finished, again judging from this paper, on the 6th of May; that paper was received at APJ on the 15th. And you thank Helen Czaplicki for typing the manuscript under great pressure. Q know Helen is just great) Why such a hurry?

Sandage:

It was a tremendously exciting time. One can't really reconstruct the electricity in the atmosphere. It was, I suppose, a time when so many of these things were coming together, from all different groups.

Weart:

Was there a feeling you might be scooped?

Sandage:

No, I don't think so. It was a momentum of events that built, and I suppose the carrying of that momentum forward caused it all to have to come out.

Weart:

The excitement of it.

Sandage:

The excitement was other-worldly.

Weart:

Then, I was also interested in a paper with the Burbidges, where you talk about the fact that these things can't last very long.

Sandage:

In THE REVIEWS OF MODERN PHYSICS.^[28]

Weart:

Was that it? Anywayt that a strong radio source can only last a million years or something like that. Then in the paper you go perhaps a little beyond that to suggest that these "quasi —stellar galaxies" are a normal phase in the life of galaxies. What's the origin of this? This is going a step farther.

Sandage:

I can't remember the sequence of eventsy but it was believed by a number of people that because these objects had red shiftsy they were in fact connected with galaxies. That was a generalization, because the only thing that was known that had large red shifts was galaxies.

Weart:

Right. In fact you called these things quasi-stellar galaxies——

Sandage:

Yes.

Weart:

—although no nebulosity had been seen ......

Weart:

Not at that time.

Weart:

Was it just because, if it was that far away it had to be a galaxy or you wouldn't see it, that sort of thing?

Sandage:

I suppose it's right. I can't reconstruct what was known and speculated on at that time, but the possibility was clear enough that they were events in the nuclei of galaxies at thattime, that that seemed to be the most reasonable explanation for the bizarreness.

Weart:

In the nuclei of galaxies?

Sandage:

Did we say that in the paper?

Weart:

No. You just said, a normal "phase" in the life of galaxies. It’s very plausible, and yet—to me, in 1950 the normal galaxy was seen as a very stately object indeed. First you have them collapse, then you have the M82 explosion, and now you make it that the average galaxy in fact has an explosive phase. It's quite a shift.

Sandage:

Yes it was. Part of that change in climate came about through a rediscussion by Lo Walter of the Seyfert phenomenon. The subject of the Seyfert galaxies lay dormant for 20 years, and in this period, perhaps 1950, 1952-53, he resurrected the subject, and understood from the optical surface brightness that something enormous was happening. This was before the radio connectiony before any of the quasars were found. That's a fundamental paper in the APJ. He used the direct plates of the Mt. Wilson Observatory and made an assessment which is the beginning of the modern work on Seyferts.

Weart:

I see, so you had discussed this and this was sort of in the back of your mind?

Sandage:

I had refereed that paper.

Weart:

Oh, I see.

Sandage:

Also, the plates are in the files here. He came to Pasadena, and I got the plates out, because—

Weart:

Oh, you got the plates for him?

Sandage:

The plates are in Hubble's files. These are the file,.ii here (pointing to files in corner of office).

Weart:

I see. Those are locked, by the way. What's in those.

Sandage:

It's all the 200-inch plates, taken to complete the photography of the Shapley-Ames (catalog) up to the present. So it's an amalgam of Made's plates, Humason's platesy Minkowski’s plates, and my plates.

Weart:

I see. That's a lot of telescope time. OK, after this paper came out, it was challenged. I wonder if you could talk about that?

Sandage:

Yes. We left for England right awayr and then it was challenged by Kinman and by Lynds.

Weart:

You left for England?

Sandage:

I spent three months in England working with Olin Eggen on materials we brought back from South Africa. I went to South Africa in 1958 with Eggen on an astronomical expedition.

Weart:

So how did you first become aware of these challenges?

Sandage:

Lynds sent me a copy of the presentation. I was to give a paper at the third Texas Symposiumy and instead of flying back to England, I decided not to do that, and asked Roger Lynds to tell about various things. He did, and it was the analysis of the distribution of halo white dwarfs, and why all of the objects claimed to be quasars were not quasars.

Weart:

You had asked him to talk about this?

Sandage:

About some aspect of the quasar distribution. He also had his new image tube spectrograph, which was the most efficient ypectrograph then in use, and was beginning to make observations.

Weart:

You asked him to talk in your stead; because you couldn't do, you said, "Here, you go and talk about something," and then he chose to talk about this?

Sandage:

Yes.

Weart:

I see. He sent you a copy of the paper.

Sandage:

He's a free and open person. Everything is up front. Then Kinman had been doing RR Lyrae star distributions in the halo, and he believed they would imitate the color distribution ...... the RR Lyrae star density gave a density of horizontal branch stars which would imitate, he believed, the distribution we'd found.

Weart:

Entirely, even the very faint ones.

Sandage:

That's what he claimed. They couldn't challenge the spectra that had been obtained, but the red shifts were a very small fraction of the total. So it was then believed that all this was absolutely premature and none of it was right. For three years or so, the question of whether there were appreciable numbers of radio-quiet quasars was up in the air. It was at that time that Luyten said, "Let's make another survey." So we began a program of six survey fields in different galactic latitudes, for blue objects, photometry and spectra. Three papers have appeared—one by Schmidt, and two by Luyten and myself. Now two of the survey fields have been completed, almost entirely, so the surface density or radio-quiet quasars is now pretty well established to (magnitude) 18-1/2. But it's taken all the time from 1963 to the present, the last run at Palomar was my final spectroscopic run on the 15 hour field. It's quite true, there are white dwarf interlopers or noise.. They begin to disappear at 17. The radio-quiet quasar population comes in with a vengence at 17 and dominates at 19. There are more quasars per square degree at 18 1/2—by a factor of five—than was claimed in the discovery paper. But Herbig's dictum is right: one should have waited for ten years to understand the situation, before claiming so much.

Weart:

I see. Is your general feeling, when you get into a controversy, to sort of back off and do a thorough study of it?

Sandage:

Well, that's the only way one can solve the problem. :1 think rhetoric and debate, at particular times, are useless.

Weart:

So you won't participate.

Sandage:

I won't participate in debates on scientific matters where the answer requires more data. It leads to the question of debating style and persuasiveness and the audience, when there is an answer, that you can get only by going back to the telescope.

Weart:

OK. Well, that's probably a good place to stop for today. It's 5:20.

Sandage:

OK.